@c DO NOT EDIT!  Generated automatically by munge-texi.pl.

@c Copyright (C) 1996-2015 John W. Eaton
@c
@c This file is part of Octave.
@c
@c Octave is free software; you can redistribute it and/or modify it
@c under the terms of the GNU General Public License as published by the
@c Free Software Foundation; either version 3 of the License, or (at
@c your option) any later version.
@c
@c Octave is distributed in the hope that it will be useful, but WITHOUT
@c ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
@c FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
@c for more details.
@c
@c You should have received a copy of the GNU General Public License
@c along with Octave; see the file COPYING.  If not, see
@c <http://www.gnu.org/licenses/>.

@node Statistics
@chapter Statistics

Octave has support for various statistical methods.  This includes
basic descriptive statistics, probability distributions, statistical tests,
random number generation, and much more.

The functions that analyze data all assume that multi-dimensional data
is arranged in a matrix where each row is an observation, and each
column is a variable.  Thus, the matrix defined by

@example
@group
a = [ 0.9, 0.7;
      0.1, 0.1;
      0.5, 0.4 ];
@end group
@end example

@noindent
contains three observations from a two-dimensional distribution.
While this is the default data arrangement, most functions support
different arrangements.

It should be noted that the statistics functions don't test for data
containing NaN, NA, or Inf.  These values need to be detected and dealt
with explicitly.  See @ref{XREFisnan,,isnan}, @ref{XREFisna,,isna},
@ref{XREFisinf,,isinf}, @ref{XREFisfinite,,isfinite}.

@menu
* Descriptive Statistics::
* Basic Statistical Functions::
* Statistical Plots::
* Correlation and Regression Analysis::
* Distributions::
* Tests::
* Random Number Generation::
@end menu

@node Descriptive Statistics
@section Descriptive Statistics

One principal goal of descriptive statistics is to represent the essence of a
large data set concisely.  Octave provides the mean, median, and mode functions
which all summarize a data set with just a single number corresponding to
the central tendency of the data.

@c mean scripts/statistics/base/mean.m
@anchor{XREFmean}
@deftypefn  {Function File} {} mean (@var{x})
@deftypefnx {Function File} {} mean (@var{x}, @var{dim})
@deftypefnx {Function File} {} mean (@var{x}, @var{opt})
@deftypefnx {Function File} {} mean (@var{x}, @var{dim}, @var{opt})
Compute the mean of the elements of the vector @var{x}.

The mean is defined as

@tex
$$ {\rm mean}(x) = \bar{x} = {1\over N} \sum_{i=1}^N x_i $$
@end tex
@ifnottex

@example
mean (x) = SUM_i x(i) / N
@end example

@end ifnottex
If @var{x} is a matrix, compute the mean for each column and return them
in a row vector.

If the optional argument @var{dim} is given, operate along this dimension.

The optional argument @var{opt} selects the type of mean to compute.
The following options are recognized:

@table @asis
@item @qcode{"a"}
Compute the (ordinary) arithmetic mean.  [default]

@item @qcode{"g"}
Compute the geometric mean.

@item @qcode{"h"}
Compute the harmonic mean.
@end table

Both @var{dim} and @var{opt} are optional.  If both are supplied, either
may appear first.
@seealso{@ref{XREFmedian,,median}, @ref{XREFmode,,mode}}
@end deftypefn


@c median scripts/statistics/base/median.m
@anchor{XREFmedian}
@deftypefn  {Function File} {} median (@var{x})
@deftypefnx {Function File} {} median (@var{x}, @var{dim})
Compute the median value of the elements of the vector @var{x}.

When the elements of @var{x} are sorted, the median is defined as
@tex
$$
{\rm median} (x) =
  \cases{x(\lceil N/2\rceil), & $N$ odd;\cr
          (x(N/2)+x(N/2+1))/2, & $N$ even.}
$$
@end tex
@ifnottex

@example
@group
              x(ceil(N/2))             N odd
median (x) =
             (x(N/2) + x((N/2)+1))/2   N even
@end group
@end example

@end ifnottex
If @var{x} is a matrix, compute the median value for each column and
return them in a row vector.

If the optional @var{dim} argument is given, operate along this dimension.
@seealso{@ref{XREFmean,,mean}, @ref{XREFmode,,mode}}
@end deftypefn


@c mode scripts/statistics/base/mode.m
@anchor{XREFmode}
@deftypefn  {Function File} {} mode (@var{x})
@deftypefnx {Function File} {} mode (@var{x}, @var{dim})
@deftypefnx {Function File} {[@var{m}, @var{f}, @var{c}] =} mode (@dots{})
Compute the most frequently occurring value in a dataset (mode).

@code{mode} determines the frequency of values along the first non-singleton
dimension and returns the value with the highest frequency.  If two, or
more, values have the same frequency @code{mode} returns the smallest.

If the optional argument @var{dim} is given, operate along this dimension.

The return variable @var{f} is the number of occurrences of the mode in
the dataset.

The cell array @var{c} contains all of the elements with the maximum
frequency.
@seealso{@ref{XREFmean,,mean}, @ref{XREFmedian,,median}}
@end deftypefn


Using just one number, such as the mean, to represent an entire data set may
not give an accurate picture of the data.  One way to characterize the fit is
to measure the dispersion of the data.  Octave provides several functions for
measuring dispersion.

@c range scripts/statistics/base/range.m
@anchor{XREFrange}
@deftypefn  {Function File} {} range (@var{x})
@deftypefnx {Function File} {} range (@var{x}, @var{dim})
Return the range, i.e., the difference between the maximum and the minimum
of the input data.

If @var{x} is a vector, the range is calculated over the elements of
@var{x}.  If @var{x} is a matrix, the range is calculated over each column
of @var{x}.

If the optional argument @var{dim} is given, operate along this dimension.

The range is a quickly computed measure of the dispersion of a data set, but
is less accurate than @code{iqr} if there are outlying data points.
@seealso{@ref{XREFiqr,,iqr}, @ref{XREFstd,,std}}
@end deftypefn


@c iqr scripts/statistics/base/iqr.m
@anchor{XREFiqr}
@deftypefn  {Function File} {} iqr (@var{x})
@deftypefnx {Function File} {} iqr (@var{x}, @var{dim})
Return the interquartile range, i.e., the difference between the upper
and lower quartile of the input data.

If @var{x} is a matrix, do the above for first non-singleton dimension of
@var{x}.

If the optional argument @var{dim} is given, operate along this dimension.

As a measure of dispersion, the interquartile range is less affected by
outliers than either @code{range} or @code{std}.
@seealso{@ref{XREFrange,,range}, @ref{XREFstd,,std}}
@end deftypefn


@c meansq scripts/statistics/base/meansq.m
@anchor{XREFmeansq}
@deftypefn  {Function File} {} meansq (@var{x})
@deftypefnx {Function File} {} meansq (@var{x}, @var{dim})
Compute the mean square of the elements of the vector @var{x}.

The mean square is defined as
@tex
$$
{\rm meansq} (x) = {\sum_{i=1}^N {x_i}^2 \over N}
$$
where $\bar{x}$ is the mean value of $x$.
@end tex
@ifnottex

@example
@group
meansq (x) = 1/N SUM_i x(i)^2
@end group
@end example

@end ifnottex
For matrix arguments, return a row vector containing the mean square
of each column.

If the optional argument @var{dim} is given, operate along this dimension.
@seealso{@ref{XREFvar,,var}, @ref{XREFstd,,std}, @ref{XREFmoment,,moment}}
@end deftypefn


@c std scripts/statistics/base/std.m
@anchor{XREFstd}
@deftypefn  {Function File} {} std (@var{x})
@deftypefnx {Function File} {} std (@var{x}, @var{opt})
@deftypefnx {Function File} {} std (@var{x}, @var{opt}, @var{dim})
Compute the standard deviation of the elements of the vector @var{x}.

The standard deviation is defined as
@tex
$$
{\rm std} (x) = \sigma = \sqrt{{\sum_{i=1}^N (x_i - \bar{x})^2 \over N - 1}}
$$
where $\bar{x}$ is the mean value of $x$ and $N$ is the number of elements.
@end tex
@ifnottex

@example
@group
std (x) = sqrt ( 1/(N-1) SUM_i (x(i) - mean(x))^2 )
@end group
@end example

@noindent
where @math{N} is the number of elements.
@end ifnottex

If @var{x} is a matrix, compute the standard deviation for each column and
return them in a row vector.

The argument @var{opt} determines the type of normalization to use.
Valid values are

@table @asis
@item 0:
  normalize with @math{N-1}, provides the square root of the best unbiased
estimator of the variance [default]

@item 1:
  normalize with @math{N}, this provides the square root of the second
moment around the mean
@end table

If the optional argument @var{dim} is given, operate along this dimension.
@seealso{@ref{XREFvar,,var}, @ref{XREFrange,,range}, @ref{XREFiqr,,iqr}, @ref{XREFmean,,mean}, @ref{XREFmedian,,median}}
@end deftypefn


In addition to knowing the size of a dispersion it is useful to know the shape
of the data set.  For example, are data points massed to the left or right
of the mean?  Octave provides several common measures to describe the shape
of the data set.  Octave can also calculate moments allowing arbitrary shape
measures to be developed.

@c var scripts/statistics/base/var.m
@anchor{XREFvar}
@deftypefn  {Function File} {} var (@var{x})
@deftypefnx {Function File} {} var (@var{x}, @var{opt})
@deftypefnx {Function File} {} var (@var{x}, @var{opt}, @var{dim})
Compute the variance of the elements of the vector @var{x}.

The variance is defined as
@tex
$$
{\rm var} (x) = \sigma^2 = {\sum_{i=1}^N (x_i - \bar{x})^2 \over N - 1}
$$
where $\bar{x}$ is the mean value of $x$.
@end tex
@ifnottex

@example
@group
var (x) = 1/(N-1) SUM_i (x(i) - mean(x))^2
@end group
@end example

@end ifnottex
If @var{x} is a matrix, compute the variance for each column and return
them in a row vector.

The argument @var{opt} determines the type of normalization to use.
Valid values are

@table @asis
@item 0:
  normalize with @math{N-1}, provides the best unbiased estimator of the
variance [default]

@item 1:
  normalizes with @math{N}, this provides the second moment around the mean
@end table

If @math{N==1} the value of @var{opt} is ignored and normalization by
@math{N} is used.

If the optional argument @var{dim} is given, operate along this dimension.
@seealso{@ref{XREFcov,,cov}, @ref{XREFstd,,std}, @ref{XREFskewness,,skewness}, @ref{XREFkurtosis,,kurtosis}, @ref{XREFmoment,,moment}}
@end deftypefn


@c skewness scripts/statistics/base/skewness.m
@anchor{XREFskewness}
@deftypefn  {Function File} {} skewness (@var{x})
@deftypefnx {Function File} {} skewness (@var{x}, @var{flag})
@deftypefnx {Function File} {} skewness (@var{x}, @var{flag}, @var{dim})
Compute the sample skewness of the elements of @var{x}.

The sample skewness is defined as
@tex
$$
{\rm skewness} (@var{x}) = {{{1\over N}\,
         \sum_{i=1}^N (@var{x}_i - \bar{@var{x}})^3} \over \sigma^3},
$$
where $N$ is the length of @var{x}, $\bar{@var{x}}$ its mean and $\sigma$
its (uncorrected) standard deviation.
@end tex
@ifnottex

@example
@group
               mean ((@var{x} - mean (@var{x})).^3)
skewness (@var{X}) = ------------------------.
                      std (@var{x}).^3
@end group
@end example

@end ifnottex

@noindent
The optional argument @var{flag} controls which normalization is used.
If @var{flag} is equal to 1 (default value, used when @var{flag} is omitted
or empty), return the sample skewness as defined above.  If @var{flag} is
equal to 0, return the adjusted skewness coefficient instead:
@tex
$$
{\rm skewness} (@var{x}) = {\sqrt{N (N - 1)} \over N - 2} \times \,
  {{{1 \over N} \sum_{i=1}^N (@var{x}_i - \bar{@var{x}})^3} \over \sigma^3}
$$
@end tex
@ifnottex

@example
@group
                  sqrt (N*(N-1))   mean ((@var{x} - mean (@var{x})).^3)
skewness (@var{X}, 0) = -------------- * ------------------------.
                      (N - 2)             std (@var{x}).^3
@end group
@end example

@end ifnottex
The adjusted skewness coefficient is obtained by replacing the sample second
and third central moments by their bias-corrected versions.

If @var{x} is a matrix, or more generally a multi-dimensional array, return
the skewness along the first non-singleton dimension.  If the optional
@var{dim} argument is given, operate along this dimension.

@seealso{@ref{XREFvar,,var}, @ref{XREFkurtosis,,kurtosis}, @ref{XREFmoment,,moment}}
@end deftypefn


@c kurtosis scripts/statistics/base/kurtosis.m
@anchor{XREFkurtosis}
@deftypefn  {Function File} {} kurtosis (@var{x})
@deftypefnx {Function File} {} kurtosis (@var{x}, @var{flag})
@deftypefnx {Function File} {} kurtosis (@var{x}, @var{flag}, @var{dim})
Compute the sample kurtosis of the elements of @var{x}.

The sample kurtosis is defined as
@tex
$$
\kappa_1 = {{{1\over N}\,
         \sum_{i=1}^N (@var{x}_i - \bar{@var{x}})^4} \over \sigma^4},
$$
where $N$ is the length of @var{x}, $\bar{@var{x}}$ its mean, and $\sigma$
its (uncorrected) standard deviation.
@end tex
@ifnottex

@example
@group
     mean ((@var{x} - mean (@var{x})).^4)
k1 = ------------------------
            std (@var{x}).^4
@end group
@end example

@end ifnottex

@noindent
The optional argument @var{flag} controls which normalization is used.
If @var{flag} is equal to 1 (default value, used when @var{flag} is omitted
or empty), return the sample kurtosis as defined above.  If @var{flag} is
equal to 0, return the @w{"bias-corrected"} kurtosis coefficient instead:
@tex
$$
\kappa_0 = 3 + {\scriptstyle N - 1 \over \scriptstyle (N - 2)(N - 3)} \,
    \left( (N + 1)\, \kappa_1 - 3 (N - 1) \right)
$$
@end tex
@ifnottex

@example
@group
              N - 1
k0 = 3 + -------------- * ((N + 1) * k1 - 3 * (N - 1))
         (N - 2)(N - 3)
@end group
@end example

@end ifnottex
The bias-corrected kurtosis coefficient is obtained by replacing the sample
second and fourth central moments by their unbiased versions.  It is an
unbiased estimate of the population kurtosis for normal populations.

If @var{x} is a matrix, or more generally a multi-dimensional array, return
the kurtosis along the first non-singleton dimension.  If the optional
@var{dim} argument is given, operate along this dimension.

@seealso{@ref{XREFvar,,var}, @ref{XREFskewness,,skewness}, @ref{XREFmoment,,moment}}
@end deftypefn


@c moment scripts/statistics/base/moment.m
@anchor{XREFmoment}
@deftypefn  {Function File} {} moment (@var{x}, @var{p})
@deftypefnx {Function File} {} moment (@var{x}, @var{p}, @var{type})
@deftypefnx {Function File} {} moment (@var{x}, @var{p}, @var{dim})
@deftypefnx {Function File} {} moment (@var{x}, @var{p}, @var{type}, @var{dim})
@deftypefnx {Function File} {} moment (@var{x}, @var{p}, @var{dim}, @var{type})
Compute the @var{p}-th central moment of the vector @var{x}.

@tex
$$
{\sum_{i=1}^N (x_i - \bar{x})^p \over N}
$$
@end tex
@ifnottex

@example
@group
1/N SUM_i (x(i) - mean(x))^p
@end group
@end example

@end ifnottex

If @var{x} is a matrix, return the row vector containing the @var{p}-th
central moment of each column.

If the optional argument @var{dim} is given, operate along this dimension.

The optional string @var{type} specifies the type of moment to be computed.
Valid options are:

@table @asis
@item @qcode{"c"}
  Central Moment (default).

@item  @qcode{"a"}
@itemx @qcode{"ac"}
  Absolute Central Moment.  The moment about the mean ignoring sign
defined as
@tex
$$
{\sum_{i=1}^N {\left| x_i - \bar{x} \right|}^p \over N}
$$
@end tex
@ifnottex

@example
@group
1/N SUM_i (abs (x(i) - mean(x)))^p
@end group
@end example

@end ifnottex

@item @qcode{"r"}
  Raw Moment.  The moment about zero defined as

@tex
$$
{\rm moment} (x) = { \sum_{i=1}^N {x_i}^p \over N }
$$
@end tex
@ifnottex

@example
@group
moment (x) = 1/N SUM_i x(i)^p
@end group
@end example

@end ifnottex

@item @nospell{@qcode{"ar"}}
  Absolute Raw Moment.  The moment about zero ignoring sign defined as
@tex
$$
{\sum_{i=1}^N {\left| x_i \right|}^p \over N}
$$
@end tex
@ifnottex

@example
@group
1/N SUM_i ( abs (x(i)) )^p
@end group
@end example

@end ifnottex
@end table

If both @var{type} and @var{dim} are given they may appear in any order.
@seealso{@ref{XREFvar,,var}, @ref{XREFskewness,,skewness}, @ref{XREFkurtosis,,kurtosis}}
@end deftypefn


@c quantile scripts/statistics/base/quantile.m
@anchor{XREFquantile}
@deftypefn  {Function File} {@var{q} =} quantile (@var{x})
@deftypefnx {Function File} {@var{q} =} quantile (@var{x}, @var{p})
@deftypefnx {Function File} {@var{q} =} quantile (@var{x}, @var{p}, @var{dim})
@deftypefnx {Function File} {@var{q} =} quantile (@var{x}, @var{p}, @var{dim}, @var{method})
For a sample, @var{x}, calculate the quantiles, @var{q}, corresponding to
the cumulative probability values in @var{p}.  All non-numeric values (NaNs)
of @var{x} are ignored.

If @var{x} is a matrix, compute the quantiles for each column and
return them in a matrix, such that the i-th row of @var{q} contains
the @var{p}(i)th quantiles of each column of @var{x}.

If @var{p} is unspecified, return the quantiles for
@code{[0.00 0.25 0.50 0.75 1.00]}.
The optional argument @var{dim} determines the dimension along which
the quantiles are calculated.  If @var{dim} is omitted it defaults to
the first non-singleton dimension.

The methods available to calculate sample quantiles are the nine methods
used by R (@url{http://www.r-project.org/}).  The default value is
@w{METHOD = 5}.

Discontinuous sample quantile methods 1, 2, and 3

@enumerate 1
@item Method 1: Inverse of empirical distribution function.

@item Method 2: Similar to method 1 but with averaging at discontinuities.

@item Method 3: SAS definition: nearest even order statistic.
@end enumerate

Continuous sample quantile methods 4 through 9, where p(k) is the linear
interpolation function respecting each methods' representative cdf.

@enumerate 4
@item Method 4: p(k) = k / n.  That is, linear interpolation of the
empirical cdf.

@item Method 5: p(k) = (k - 0.5) / n.  That is a piecewise linear function
where the knots are the values midway through the steps of the empirical
cdf.

@item Method 6: p(k) = k / (n + 1).

@item Method 7: p(k) = (k - 1) / (n - 1).

@item Method 8: p(k) = (k - 1/3) / (n + 1/3).  The resulting quantile
estimates are approximately median-unbiased regardless of the distribution
of @var{x}.

@item Method 9: p(k) = (k - 3/8) / (n + 1/4).  The resulting quantile
estimates are approximately unbiased for the expected order statistics if
@var{x} is normally distributed.
@end enumerate

@nospell{Hyndman and Fan} (1996) recommend method 8.  Maxima, S, and R
(versions prior to 2.0.0) use 7 as their default.  Minitab and SPSS
use method 6.  @sc{matlab} uses method 5.

References:

@itemize @bullet
@item @nospell{Becker, R. A., Chambers, J. M. and Wilks, A. R.} (1988)
The New S Language.  Wadsworth & Brooks/Cole.

@item @nospell{Hyndman, R. J. and Fan, Y.} (1996) Sample quantiles in
statistical packages, American Statistician, 50, 361--365.

@item R: A Language and Environment for Statistical Computing;
@url{http://cran.r-project.org/doc/manuals/fullrefman.pdf}.
@end itemize

Examples:
@c Set example in small font to prevent overfull line

@smallexample
@group
x = randi (1000, [10, 1]);  # Create empirical data in range 1-1000
q = quantile (x, [0, 1]);   # Return minimum, maximum of distribution
q = quantile (x, [0.25 0.5 0.75]); # Return quartiles of distribution
@end group
@end smallexample
@seealso{@ref{XREFprctile,,prctile}}
@end deftypefn


@c prctile scripts/statistics/base/prctile.m
@anchor{XREFprctile}
@deftypefn  {Function File} {@var{q} =} prctile (@var{x})
@deftypefnx {Function File} {@var{q} =} prctile (@var{x}, @var{p})
@deftypefnx {Function File} {@var{q} =} prctile (@var{x}, @var{p}, @var{dim})
For a sample @var{x}, compute the quantiles, @var{q}, corresponding
to the cumulative probability values, @var{p}, in percent.

If @var{x} is a matrix, compute the percentiles for each column and return
them in a matrix, such that the i-th row of @var{y} contains the
@var{p}(i)th percentiles of each column of @var{x}.

If @var{p} is unspecified, return the quantiles for @code{[0 25 50 75 100]}.

The optional argument @var{dim} determines the dimension along which the
percentiles are calculated.  If @var{dim} is omitted it defaults to the
first non-singleton dimension.

Programming Note: All non-numeric values (NaNs) of @var{x} are ignored.
@seealso{@ref{XREFquantile,,quantile}}
@end deftypefn


A summary view of a data set can be generated quickly with the
@code{statistics} function.

@c statistics scripts/statistics/base/statistics.m
@anchor{XREFstatistics}
@deftypefn  {Function File} {} statistics (@var{x})
@deftypefnx {Function File} {} statistics (@var{x}, @var{dim})
Return a vector with the minimum, first quartile, median, third quartile,
maximum, mean, standard deviation, skewness, and kurtosis of the elements of
the vector @var{x}.

If @var{x} is a matrix, calculate statistics over the first non-singleton
dimension.

If the optional argument @var{dim} is given, operate along this dimension.
@seealso{@ref{XREFmin,,min}, @ref{XREFmax,,max}, @ref{XREFmedian,,median}, @ref{XREFmean,,mean}, @ref{XREFstd,,std}, @ref{XREFskewness,,skewness}, @ref{XREFkurtosis,,kurtosis}}
@end deftypefn


@node Basic Statistical Functions
@section Basic Statistical Functions

Octave supports various helpful statistical functions.  Many are useful as
initial steps to prepare a data set for further analysis.  Others provide
different measures from those of the basic descriptive statistics.

@c center scripts/statistics/base/center.m
@anchor{XREFcenter}
@deftypefn  {Function File} {} center (@var{x})
@deftypefnx {Function File} {} center (@var{x}, @var{dim})
Center data by subtracting its mean.

If @var{x} is a vector, subtract its mean.

If @var{x} is a matrix, do the above for each column.

If the optional argument @var{dim} is given, operate along this dimension.

Programming Note: @code{center} has obvious application for normalizing
statistical data.  It is also useful for improving the precision of general
numerical calculations.  Whenever there is a large value that is common
to a batch of data, the mean can be subtracted off, the calculation
performed, and then the mean added back to obtain the final answer.
@seealso{@ref{XREFzscore,,zscore}}
@end deftypefn


@c zscore scripts/statistics/base/zscore.m
@anchor{XREFzscore}
@deftypefn  {Function File} {@var{z} =} zscore (@var{x})
@deftypefnx {Function File} {@var{z} =} zscore (@var{x}, @var{opt})
@deftypefnx {Function File} {@var{z} =} zscore (@var{x}, @var{opt}, @var{dim})
@deftypefnx {Function File} {[@var{z}, @var{mu}, @var{sigma}] =} zscore (@dots{})
Compute the Z score of @var{x}

If @var{x} is a vector, subtract its mean and divide by its standard
deviation.  If the standard deviation is zero, divide by 1 instead.

The optional parameter @var{opt} determines the normalization to use when
computing the standard deviation and has the same definition as the
corresponding parameter for @code{std}.

If @var{x} is a matrix, calculate along the first non-singleton dimension.
If the third optional argument @var{dim} is given, operate along this
dimension.

The optional outputs @var{mu} and @var{sigma} contain the mean and standard
deviation.

@seealso{@ref{XREFmean,,mean}, @ref{XREFstd,,std}, @ref{XREFcenter,,center}}
@end deftypefn


@c histc scripts/statistics/base/histc.m
@anchor{XREFhistc}
@deftypefn  {Function File} {@var{n} =} histc (@var{x}, @var{edges})
@deftypefnx {Function File} {@var{n} =} histc (@var{x}, @var{edges}, @var{dim})
@deftypefnx {Function File} {[@var{n}, @var{idx}] =} histc (@dots{})
Compute histogram counts.

When @var{x} is a vector, the function counts the number of elements of
@var{x} that fall in the histogram bins defined by @var{edges}.  This must be
a vector of monotonically increasing values that define the edges of the
histogram bins.  @code{@var{n}(k)} contains the number of elements in
@var{x} for which @code{@var{edges}(k) <= @var{x} < @var{edges}(k+1)}.
The final element of @var{n} contains the number of elements of @var{x}
exactly equal to the last element of @var{edges}.

When @var{x} is an @math{N}-dimensional array, the computation is carried
out along dimension @var{dim}.  If not specified @var{dim} defaults to the
first non-singleton dimension.

When a second output argument is requested an index matrix is also returned.
The @var{idx} matrix has the same size as @var{x}.  Each element of @var{idx}
contains the index of the histogram bin in which the corresponding element
of @var{x} was counted.
@seealso{@ref{XREFhist,,hist}}
@end deftypefn


@c FIXME: really want to put a reference to unique here
@c @DOCSTRING(values)

@c nchoosek scripts/specfun/nchoosek.m
@anchor{XREFnchoosek}
@deftypefn  {Function File} {@var{c} =} nchoosek (@var{n}, @var{k})
@deftypefnx {Function File} {@var{c} =} nchoosek (@var{set}, @var{k})

Compute the binomial coefficient of @var{n} or list all possible
combinations of a @var{set} of items.

If @var{n} is a scalar then calculate the binomial coefficient
of @var{n} and @var{k} which is defined as
@tex
$$
 {n \choose k} = {n (n-1) (n-2) \cdots (n-k+1) \over k!}
               = {n! \over k! (n-k)!}
$$
@end tex
@ifnottex

@example
@group
 /   \
 | n |    n (n-1) (n-2) @dots{} (n-k+1)       n!
 |   |  = ------------------------- =  ---------
 | k |               k!                k! (n-k)!
 \   /
@end group
@end example

@end ifnottex
@noindent
This is the number of combinations of @var{n} items taken in groups of
size @var{k}.

If the first argument is a vector, @var{set}, then generate all
combinations of the elements of @var{set}, taken @var{k} at a time, with
one row per combination.  The result @var{c} has @var{k} columns and
@w{@code{nchoosek (length (@var{set}), @var{k})}} rows.

For example:

How many ways can three items be grouped into pairs?

@example
@group
nchoosek (3, 2)
   @result{} 3
@end group
@end example

What are the possible pairs?

@example
@group
nchoosek (1:3, 2)
   @result{}  1   2
       1   3
       2   3
@end group
@end example

Programming Note: When calculating the binomial coefficient @code{nchoosek}
works only for non-negative, integer arguments.  Use @code{bincoeff} for
non-integer and negative scalar arguments, or for computing many binomial
coefficients at once with vector inputs for @var{n} or @var{k}.

@seealso{@ref{XREFbincoeff,,bincoeff}, @ref{XREFperms,,perms}}
@end deftypefn


@c perms scripts/specfun/perms.m
@anchor{XREFperms}
@deftypefn {Function File} {} perms (@var{v})
Generate all permutations of @var{v} with one row per permutation.

The result has size @code{factorial (@var{n}) * @var{n}}, where @var{n}
is the length of @var{v}.

Example

@example
@group
perms ([1, 2, 3])
@result{}
  1   2   3
  2   1   3
  1   3   2
  2   3   1
  3   1   2
  3   2   1
@end group
@end example

Programming Note: The maximum length of @var{v} should be less than or
equal to 10 to limit memory consumption.
@seealso{@ref{XREFpermute,,permute}, @ref{XREFrandperm,,randperm}, @ref{XREFnchoosek,,nchoosek}}
@end deftypefn


@c ranks scripts/statistics/base/ranks.m
@anchor{XREFranks}
@deftypefn {Function File} {} ranks (@var{x}, @var{dim})
Return the ranks of @var{x} along the first non-singleton dimension
adjusted for ties.

If the optional argument @var{dim} is given, operate along this dimension.
@seealso{@ref{XREFspearman,,spearman}, @ref{XREFkendall,,kendall}}
@end deftypefn


@c run_count scripts/statistics/base/run_count.m
@anchor{XREFrun_count}
@deftypefn  {Function File} {} run_count (@var{x}, @var{n})
@deftypefnx {Function File} {} run_count (@var{x}, @var{n}, @var{dim})
Count the upward runs along the first non-singleton dimension of @var{x}
of length 1, 2, @dots{}, @var{n}-1 and greater than or equal to @var{n}.

If the optional argument @var{dim} is given then operate along this
dimension.
@seealso{@ref{XREFrunlength,,runlength}}
@end deftypefn


@c runlength scripts/statistics/base/runlength.m
@anchor{XREFrunlength}
@deftypefn  {Function File} {count =} runlength (@var{x})
@deftypefnx {Function File} {[count, value] =} runlength (@var{x})
Find the lengths of all sequences of common values.

@var{count} is a vector with the lengths of each repeated value.

The optional output @var{value} contains the value that was repeated in
the sequence.

@example
@group
runlength ([2, 2, 0, 4, 4, 4, 0, 1, 1, 1, 1])
@result{}  [2, 1, 3, 1, 4]
@end group
@end example
@seealso{@ref{XREFrun_count,,run_count}}
@end deftypefn


@c probit scripts/statistics/base/probit.m
@anchor{XREFprobit}
@deftypefn {Function File} {} probit (@var{p})
Return the probit (the quantile of the standard normal distribution) for
each element of @var{p}.
@seealso{@ref{XREFlogit,,logit}}
@end deftypefn


@c logit scripts/statistics/base/logit.m
@anchor{XREFlogit}
@deftypefn {Function File} {} logit (@var{p})
Compute the logit for each value of @var{p}

The logit is defined as
@tex
$$
{\rm logit}(p) = \log\Big({p \over 1-p}\Big)
$$
@end tex
@ifnottex

@example
logit (@var{p}) = log (@var{p} / (1-@var{p}))
@end example

@end ifnottex
@seealso{@ref{XREFprobit,,probit}, @ref{XREFlogistic_cdf,,logistic_cdf}}
@end deftypefn


@c cloglog scripts/statistics/base/cloglog.m
@anchor{XREFcloglog}
@deftypefn {Function File} {} cloglog (@var{x})
Return the complementary log-log function of @var{x}.

The complementary log-log function is defined as
@tex
$$
{\rm cloglog}(x) = - \log (- \log (x))
$$
@end tex
@ifnottex

@example
cloglog (x) = - log (- log (@var{x}))
@end example

@end ifnottex
@end deftypefn


@c mahalanobis scripts/statistics/base/mahalanobis.m
@anchor{XREFmahalanobis}
@deftypefn {Function File} {} mahalanobis (@var{x}, @var{y})
Return the Mahalanobis' D-square distance between the multivariate
samples @var{x} and @var{y}.

The data @var{x} and @var{y} must have the same number of components
(columns), but may have a different number of observations (rows).
@end deftypefn


@c table scripts/statistics/base/table.m
@anchor{XREFtable}
@deftypefn  {Function File} {[@var{t}, @var{l_x}] =} table (@var{x})
@deftypefnx {Function File} {[@var{t}, @var{l_x}, @var{l_y}] =} table (@var{x}, @var{y})
Create a contingency table @var{t} from data vectors.

The @var{l_x} and @var{l_y} vectors are the corresponding levels.

Currently, only 1- and 2-dimensional tables are supported.
@end deftypefn


@node Statistical Plots
@section Statistical Plots

@c Should hist be moved to here, or perhaps the qqplot and ppplot
@c functions should be moved to the Plotting Chapter?

Octave can create Quantile Plots (QQ-Plots), and Probability Plots
(PP-Plots).  These are simple graphical tests for determining if a
data set comes from a certain distribution.

Note that Octave can also show histograms of data
using the @code{hist} function as described in
@ref{Two-Dimensional Plots}.

@c qqplot scripts/statistics/base/qqplot.m
@anchor{XREFqqplot}
@deftypefn  {Function File} {[@var{q}, @var{s}] =} qqplot (@var{x})
@deftypefnx {Function File} {[@var{q}, @var{s}] =} qqplot (@var{x}, @var{y})
@deftypefnx {Function File} {[@var{q}, @var{s}] =} qqplot (@var{x}, @var{dist})
@deftypefnx {Function File} {[@var{q}, @var{s}] =} qqplot (@var{x}, @var{y}, @var{params})
@deftypefnx {Function File} {} qqplot (@dots{})
Perform a QQ-plot (quantile plot).

If F is the CDF of the distribution @var{dist} with parameters
@var{params} and G its inverse, and @var{x} a sample vector of length
@var{n}, the QQ-plot graphs ordinate @var{s}(@var{i}) = @var{i}-th
largest element of x versus abscissa @var{q}(@var{i}f) = G((@var{i} -
0.5)/@var{n}).

If the sample comes from F, except for a transformation of location
and scale, the pairs will approximately follow a straight line.

If the second argument is a vector @var{y} the empirical CDF of @var{y}
is used as @var{dist}.

The default for @var{dist} is the standard normal distribution.  The
optional argument @var{params} contains a list of parameters of
@var{dist}.  For example, for a quantile plot of the uniform
distribution on [2,4] and @var{x}, use

@example
qqplot (x, "unif", 2, 4)
@end example

@noindent
@var{dist} can be any string for which a function @var{distinv} or
@var{dist_inv} exists that calculates the inverse CDF of distribution
@var{dist}.

If no output arguments are given, the data are plotted directly.
@end deftypefn


@c ppplot scripts/statistics/base/ppplot.m
@anchor{XREFppplot}
@deftypefn {Function File} {[@var{p}, @var{y}] =} ppplot (@var{x}, @var{dist}, @var{params})
Perform a PP-plot (probability plot).

If F is the CDF of the distribution @var{dist} with parameters
@var{params} and @var{x} a sample vector of length @var{n}, the PP-plot
graphs ordinate @var{y}(@var{i}) = F (@var{i}-th largest element of
@var{x}) versus abscissa @var{p}(@var{i}) = (@var{i} - 0.5)/@var{n}.  If
the sample comes from F, the pairs will approximately follow a straight
line.

The default for @var{dist} is the standard normal distribution.

The optional argument @var{params} contains a list of parameters of
@var{dist}.

For example, for a probability plot of the uniform distribution on [2,4]
and @var{x}, use

@example
ppplot (x, "uniform", 2, 4)
@end example

@noindent
@var{dist} can be any string for which a function @var{dist_cdf} that
calculates the CDF of distribution @var{dist} exists.

If no output is requested then the data are plotted immediately.
@end deftypefn


@node Correlation and Regression Analysis
@section Correlation and Regression Analysis

@c FIXME: Need Intro Here

@c cov scripts/statistics/base/cov.m
@anchor{XREFcov}
@deftypefn  {Function File} {} cov (@var{x})
@deftypefnx {Function File} {} cov (@var{x}, @var{opt})
@deftypefnx {Function File} {} cov (@var{x}, @var{y})
@deftypefnx {Function File} {} cov (@var{x}, @var{y}, @var{opt})
Compute the covariance matrix.

If each row of @var{x} and @var{y} is an observation, and each column is
a variable, then the @w{(@var{i}, @var{j})-th} entry of
@code{cov (@var{x}, @var{y})} is the covariance between the @var{i}-th
variable in @var{x} and the @var{j}-th variable in @var{y}.
@tex
$$
\sigma_{ij} = {1 \over N-1} \sum_{i=1}^N (x_i - \bar{x})(y_i - \bar{y})
$$
where $\bar{x}$ and $\bar{y}$ are the mean values of $x$ and $y$.
@end tex
@ifnottex

@example
cov (x) = 1/N-1 * SUM_i (x(i) - mean(x)) * (y(i) - mean(y))
@end example

@end ifnottex

If called with one argument, compute @code{cov (@var{x}, @var{x})}, the
covariance between the columns of @var{x}.

The argument @var{opt} determines the type of normalization to use.
Valid values are

@table @asis
@item 0:
  normalize with @math{N-1}, provides the best unbiased estimator of the
covariance [default]

@item 1:
  normalize with @math{N}, this provides the second moment around the mean
@end table

Compatibility Note:: Octave always computes the covariance matrix.
For two inputs, however, @sc{matlab} will calculate
@code{cov (@var{x}(:), @var{y}(:))} whenever the number of elements in
@var{x} and @var{y} are equal.  This will result in a scalar rather than
a matrix output.  Code relying on this odd definition will need to be
changed when running in Octave.
@seealso{@ref{XREFcorr,,corr}}
@end deftypefn


@c corr scripts/statistics/base/corr.m
@anchor{XREFcorr}
@deftypefn  {Function File} {} corr (@var{x})
@deftypefnx {Function File} {} corr (@var{x}, @var{y})
Compute matrix of correlation coefficients.

If each row of @var{x} and @var{y} is an observation and each column is
a variable, then the @w{(@var{i}, @var{j})-th} entry of
@code{corr (@var{x}, @var{y})} is the correlation between the
@var{i}-th variable in @var{x} and the @var{j}-th variable in @var{y}.
@tex
$$
{\rm corr}(x,y) = {{\rm cov}(x,y) \over {\rm std}(x) {\rm std}(y)}
$$
@end tex
@ifnottex

@example
corr (x,y) = cov (x,y) / (std (x) * std (y))
@end example

@end ifnottex
If called with one argument, compute @code{corr (@var{x}, @var{x})},
the correlation between the columns of @var{x}.
@seealso{@ref{XREFcov,,cov}}
@end deftypefn


@c spearman scripts/statistics/base/spearman.m
@anchor{XREFspearman}
@deftypefn  {Function File} {} spearman (@var{x})
@deftypefnx {Function File} {} spearman (@var{x}, @var{y})
@cindex Spearman's Rho
Compute Spearman's rank correlation coefficient @var{rho}.

For two data vectors @var{x} and @var{y}, Spearman's @var{rho} is the
correlation coefficient of the ranks of @var{x} and @var{y}.

If @var{x} and @var{y} are drawn from independent distributions, @var{rho}
has zero mean and variance @code{1 / (n - 1)}, and is asymptotically
normally distributed.

@code{spearman (@var{x})} is equivalent to
@code{spearman (@var{x}, @var{x})}.
@seealso{@ref{XREFranks,,ranks}, @ref{XREFkendall,,kendall}}
@end deftypefn


@c kendall scripts/statistics/base/kendall.m
@anchor{XREFkendall}
@deftypefn  {Function File} {} kendall (@var{x})
@deftypefnx {Function File} {} kendall (@var{x}, @var{y})
@cindex Kendall's Tau
Compute Kendall's @var{tau}.

For two data vectors @var{x}, @var{y} of common length @var{n}, Kendall's
@var{tau} is the correlation of the signs of all rank differences of
@var{x} and @var{y}; i.e., if both @var{x} and @var{y} have distinct
entries, then

@tex
$$ \tau = {1 \over n(n-1)} \sum_{i,j} {\rm sign}(q_i-q_j) {\rm sign}(r_i-r_j) $$
@end tex
@ifnottex

@example
@group
         1
tau = -------   SUM sign (q(i) - q(j)) * sign (r(i) - r(j))
      n (n-1)   i,j
@end group
@end example

@end ifnottex
@noindent
in which the
@tex
$q_i$ and $r_i$
@end tex
@ifnottex
@var{q}(@var{i}) and @var{r}(@var{i})
@end ifnottex
are the ranks of @var{x} and @var{y}, respectively.

If @var{x} and @var{y} are drawn from independent distributions,
Kendall's @var{tau} is asymptotically normal with mean 0 and variance
@tex
${2 (2n+5) \over 9n(n-1)}$.
@end tex
@ifnottex
@code{(2 * (2@var{n}+5)) / (9 * @var{n} * (@var{n}-1))}.
@end ifnottex

@code{kendall (@var{x})} is equivalent to @code{kendall (@var{x},
@var{x})}.
@seealso{@ref{XREFranks,,ranks}, @ref{XREFspearman,,spearman}}
@end deftypefn


@c FIXME: Need discussion of ols & gls and references to them in optim.txi


@c logistic_regression scripts/statistics/models/logistic_regression.m
@anchor{XREFlogistic_regression}
@deftypefn {Function File} {[@var{theta}, @var{beta}, @var{dev}, @var{dl}, @var{d2l}, @var{p}] =} logistic_regression (@var{y}, @var{x}, @var{print}, @var{theta}, @var{beta})
Perform ordinal logistic regression.

Suppose @var{y} takes values in @var{k} ordered categories, and let
@code{gamma_i (@var{x})} be the cumulative probability that @var{y}
falls in one of the first @var{i} categories given the covariate
@var{x}.  Then

@example
[theta, beta] = logistic_regression (y, x)
@end example

@noindent
fits the model

@example
logit (gamma_i (x)) = theta_i - beta' * x,   i = 1 @dots{} k-1
@end example

The number of ordinal categories, @var{k}, is taken to be the number
of distinct values of @code{round (@var{y})}.  If @var{k} equals 2,
@var{y} is binary and the model is ordinary logistic regression.  The
matrix @var{x} is assumed to have full column rank.

Given @var{y} only, @code{theta = logistic_regression (y)}
fits the model with baseline logit odds only.

The full form is

@example
@group
[theta, beta, dev, dl, d2l, gamma]
   = logistic_regression (y, x, print, theta, beta)
@end group
@end example

@noindent
in which all output arguments and all input arguments except @var{y}
are optional.

Setting @var{print} to 1 requests summary information about the fitted
model to be displayed.  Setting @var{print} to 2 requests information
about convergence at each iteration.  Other values request no
information to be displayed.  The input arguments @var{theta} and
@var{beta} give initial estimates for @var{theta} and @var{beta}.

The returned value @var{dev} holds minus twice the log-likelihood.

The returned values @var{dl} and @var{d2l} are the vector of first
and the matrix of second derivatives of the log-likelihood with
respect to @var{theta} and @var{beta}.

@var{p} holds estimates for the conditional distribution of @var{y}
given @var{x}.
@end deftypefn


@node Distributions
@section Distributions

Octave has functions for computing the Probability Density Function
(PDF), the Cumulative Distribution function (CDF), and the quantile
(the inverse of the CDF) for a large number of distributions.

The following table summarizes the supported distributions (in
alphabetical order).

@tex
\vskip 6pt
{\hbox to \hsize {\hfill\vbox{\offinterlineskip \tabskip=0pt
\halign{
\vrule height2.0ex depth1.ex width 0.6pt #\tabskip=0.3em &
# \hfil & \vrule # & # \hfil & \vrule # & # \hfil & \vrule # & # \hfil &
# \vrule width 0.6pt \tabskip=0pt\cr
\noalign{\hrule height 0.6pt}
& {\bf Distribution} && {\bf PDF}      && {\bf CDF}     && {\bf Quantile}&\cr
\noalign{\hrule}
&Beta         && betapdf        && betacdf       && betainv&\cr
&Binomial     && binopdf        && binocdf       && binoinv&\cr
&Cauchy       && cauchy\_pdf    && cauchy\_cdf   && cauchy\_inv&\cr
&Chi-Square   && chi2pdf        && chi2cdf       && chi2inv&\cr
&Univariate Discrete       && discrete\_pdf  && discrete\_cdf && discrete\_inv&\cr
&Empirical    && empirical\_pdf  && empirical\_cdf && empirical\_inv&\cr
&Exponential  && exppdf         && expcdf        && expinv&\cr
&F            && fpdf           && fcdf          && finv&\cr
&Gamma        && gampdf         && gamcdf        && gaminv&\cr
&Geometric    && geopdf         && geocdf        && geoinv&\cr
&Hypergeometric  && hygepdf     && hygecdf       && hygeinv&\cr
&Kolmogorov Smirnov && {\it Not Available} && kolmogorov\_&& {\it Not Available}&\cr
&             &&                && smirnov\_cdf &&&\cr
&Laplace      && laplace\_pdf   && laplace\_cdf  && laplace\_inv&\cr
&Logistic     && logistic\_pdf  && logistic\_cdf && logistic\_inv&\cr
&Log-Normal   && lognpdf        && logncdf       && logninv&\cr
&Univariate Normal && normpdf   && normcdf       && norminv&\cr
&Pascal       && nbinpdf        && nbincdf       && nbininv&\cr
&Poisson      && poisspdf       && poisscdf      && poissinv&\cr
&Standard Normal && stdnormal\_pdf  && stdnormal\_cdf && stdnormal\_inv&\cr
&t (Student)  && tpdf           && tcdf          && tinv&\cr
&Uniform Discrete && unidpdf    && unidcdf       && unidinv&\cr
&Uniform      && unifpdf        && unifcdf       && unifinv&\cr
&Weibull      && wblpdf         && wblcdf        && wblinv&\cr
\noalign{\hrule height 0.6pt}
}}\hfill}}
@end tex
@ifnottex
@multitable @columnfractions .31 .23 .23 .23
@headitem Distribution
  @tab PDF
  @tab CDF
  @tab Quantile
@item Beta Distribution
  @tab @code{betapdf}
  @tab @code{betacdf}
  @tab @code{betainv}
@item Binomial Distribution
  @tab @code{binopdf}
  @tab @code{binocdf}
  @tab @code{binoinv}
@item Cauchy Distribution
  @tab @code{cauchy_pdf}
  @tab @code{cauchy_cdf}
  @tab @code{cauchy_inv}
@item Chi-Square Distribution
  @tab @code{chi2pdf}
  @tab @code{chi2cdf}
  @tab @code{chi2inv}
@item Univariate Discrete Distribution
  @tab @code{discrete_pdf}
  @tab @code{discrete_cdf}
  @tab @code{discrete_inv}
@item Empirical Distribution
  @tab @code{empirical_pdf}
  @tab @code{empirical_cdf}
  @tab @code{empirical_inv}
@item Exponential Distribution
  @tab @code{exppdf}
  @tab @code{expcdf}
  @tab @code{expinv}
@item F Distribution
  @tab @code{fpdf}
  @tab @code{fcdf}
  @tab @code{finv}
@item Gamma Distribution
  @tab @code{gampdf}
  @tab @code{gamcdf}
  @tab @code{gaminv}
@item Geometric Distribution
  @tab @code{geopdf}
  @tab @code{geocdf}
  @tab @code{geoinv}
@item Hypergeometric Distribution
  @tab @code{hygepdf}
  @tab @code{hygecdf}
  @tab @code{hygeinv}
@item Kolmogorov Smirnov Distribution
  @tab @emph{Not Available}
  @tab @code{kolmogorov_smirnov_cdf}
  @tab @emph{Not Available}
@item Laplace Distribution
  @tab @code{laplace_pdf}
  @tab @code{laplace_cdf}
  @tab @code{laplace_inv}
@item Logistic Distribution
  @tab @code{logistic_pdf}
  @tab @code{logistic_cdf}
  @tab @code{logistic_inv}
@item Log-Normal Distribution
  @tab @code{lognpdf}
  @tab @code{logncdf}
  @tab @code{logninv}
@item Univariate Normal Distribution
  @tab @code{normpdf}
  @tab @code{normcdf}
  @tab @code{norminv}
@item Pascal Distribution
  @tab @code{nbinpdf}
  @tab @code{nbincdf}
  @tab @code{nbininv}
@item Poisson Distribution
  @tab @code{poisspdf}
  @tab @code{poisscdf}
  @tab @code{poissinv}
@item Standard Normal Distribution
  @tab @code{stdnormal_pdf}
  @tab @code{stdnormal_cdf}
  @tab @code{stdnormal_inv}
@item t (Student) Distribution
  @tab @code{tpdf}
  @tab @code{tcdf}
  @tab @code{tinv}
@item Univariate Discrete Distribution
  @tab @code{unidpdf}
  @tab @code{unidcdf}
  @tab @code{unidinv}
@item Uniform Distribution
  @tab @code{unifpdf}
  @tab @code{unifcdf}
  @tab @code{unifinv}
@item Weibull Distribution
  @tab @code{wblpdf}
  @tab @code{wblcdf}
  @tab @code{wblinv}
@end multitable
@end ifnottex

@c betapdf scripts/statistics/distributions/betapdf.m
@anchor{XREFbetapdf}
@deftypefn {Function File} {} betapdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the Beta distribution with parameters @var{a} and @var{b}.
@end deftypefn


@c betacdf scripts/statistics/distributions/betacdf.m
@anchor{XREFbetacdf}
@deftypefn {Function File} {} betacdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the Beta distribution with parameters @var{a} and
@var{b}.
@end deftypefn


@c betainv scripts/statistics/distributions/betainv.m
@anchor{XREFbetainv}
@deftypefn {Function File} {} betainv (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the Beta distribution with parameters @var{a} and @var{b}.
@end deftypefn


@c binopdf scripts/statistics/distributions/binopdf.m
@anchor{XREFbinopdf}
@deftypefn {Function File} {} binopdf (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the binomial distribution with parameters @var{n} and @var{p},
where @var{n} is the number of trials and @var{p} is the probability of
success.
@end deftypefn


@c binocdf scripts/statistics/distributions/binocdf.m
@anchor{XREFbinocdf}
@deftypefn {Function File} {} binocdf (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the binomial distribution with parameters @var{n} and
@var{p}, where @var{n} is the number of trials and @var{p} is the
probability of success.
@end deftypefn


@c binoinv scripts/statistics/distributions/binoinv.m
@anchor{XREFbinoinv}
@deftypefn {Function File} {} binoinv (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the binomial distribution with parameters
@var{n} and @var{p}, where @var{n} is the number of trials and
@var{p} is the probability of success.
@end deftypefn


@c cauchy_pdf scripts/statistics/distributions/cauchy_pdf.m
@anchor{XREFcauchy_pdf}
@deftypefn  {Function File} {} cauchy_pdf (@var{x})
@deftypefnx {Function File} {} cauchy_pdf (@var{x}, @var{location}, @var{scale})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the Cauchy distribution with location parameter
@var{location} and scale parameter @var{scale} > 0.

Default values are @var{location} = 0, @var{scale} = 1.
@end deftypefn


@c cauchy_cdf scripts/statistics/distributions/cauchy_cdf.m
@anchor{XREFcauchy_cdf}
@deftypefn  {Function File} {} cauchy_cdf (@var{x})
@deftypefnx {Function File} {} cauchy_cdf (@var{x}, @var{location}, @var{scale})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the Cauchy distribution with location parameter
@var{location} and scale parameter @var{scale}.

Default values are @var{location} = 0, @var{scale} = 1.
@end deftypefn


@c cauchy_inv scripts/statistics/distributions/cauchy_inv.m
@anchor{XREFcauchy_inv}
@deftypefn  {Function File} {} cauchy_inv (@var{x})
@deftypefnx {Function File} {} cauchy_inv (@var{x}, @var{location}, @var{scale})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the Cauchy distribution with location parameter
@var{location} and scale parameter @var{scale}.

Default values are @var{location} = 0, @var{scale} = 1.
@end deftypefn


@c chi2pdf scripts/statistics/distributions/chi2pdf.m
@anchor{XREFchi2pdf}
@deftypefn {Function File} {} chi2pdf (@var{x}, @var{n})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the chi-square distribution with @var{n} degrees of freedom.
@end deftypefn


@c chi2cdf scripts/statistics/distributions/chi2cdf.m
@anchor{XREFchi2cdf}
@deftypefn {Function File} {} chi2cdf (@var{x}, @var{n})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the chi-square distribution with @var{n} degrees of
freedom.
@end deftypefn


@c chi2inv scripts/statistics/distributions/chi2inv.m
@anchor{XREFchi2inv}
@deftypefn {Function File} {} chi2inv (@var{x}, @var{n})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the chi-square distribution with @var{n} degrees of freedom.
@end deftypefn


@c discrete_pdf scripts/statistics/distributions/discrete_pdf.m
@anchor{XREFdiscrete_pdf}
@deftypefn {Function File} {} discrete_pdf (@var{x}, @var{v}, @var{p})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of a univariate discrete distribution which assumes the values
in @var{v} with probabilities @var{p}.
@end deftypefn


@c discrete_cdf scripts/statistics/distributions/discrete_cdf.m
@anchor{XREFdiscrete_cdf}
@deftypefn {Function File} {} discrete_cdf (@var{x}, @var{v}, @var{p})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of a univariate discrete distribution which assumes the
values in @var{v} with probabilities @var{p}.
@end deftypefn


@c discrete_inv scripts/statistics/distributions/discrete_inv.m
@anchor{XREFdiscrete_inv}
@deftypefn {Function File} {} discrete_inv (@var{x}, @var{v}, @var{p})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the univariate distribution which assumes the values in
@var{v} with probabilities @var{p}.
@end deftypefn


@c empirical_pdf scripts/statistics/distributions/empirical_pdf.m
@anchor{XREFempirical_pdf}
@deftypefn {Function File} {} empirical_pdf (@var{x}, @var{data})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the empirical distribution obtained from the
univariate sample @var{data}.
@end deftypefn


@c empirical_cdf scripts/statistics/distributions/empirical_cdf.m
@anchor{XREFempirical_cdf}
@deftypefn {Function File} {} empirical_cdf (@var{x}, @var{data})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the empirical distribution obtained from
the univariate sample @var{data}.
@end deftypefn


@c empirical_inv scripts/statistics/distributions/empirical_inv.m
@anchor{XREFempirical_inv}
@deftypefn {Function File} {} empirical_inv (@var{x}, @var{data})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the empirical distribution obtained from the
univariate sample @var{data}.
@end deftypefn


@c exppdf scripts/statistics/distributions/exppdf.m
@anchor{XREFexppdf}
@deftypefn {Function File} {} exppdf (@var{x}, @var{lambda})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the exponential distribution with mean @var{lambda}.
@end deftypefn


@c expcdf scripts/statistics/distributions/expcdf.m
@anchor{XREFexpcdf}
@deftypefn {Function File} {} expcdf (@var{x}, @var{lambda})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the exponential distribution with mean @var{lambda}.

The arguments can be of common size or scalars.
@end deftypefn


@c expinv scripts/statistics/distributions/expinv.m
@anchor{XREFexpinv}
@deftypefn {Function File} {} expinv (@var{x}, @var{lambda})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the exponential distribution with mean @var{lambda}.
@end deftypefn


@c fpdf scripts/statistics/distributions/fpdf.m
@anchor{XREFfpdf}
@deftypefn {Function File} {} fpdf (@var{x}, @var{m}, @var{n})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the F distribution with @var{m} and @var{n} degrees of freedom.
@end deftypefn


@c fcdf scripts/statistics/distributions/fcdf.m
@anchor{XREFfcdf}
@deftypefn {Function File} {} fcdf (@var{x}, @var{m}, @var{n})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the F distribution with @var{m} and @var{n} degrees of
freedom.
@end deftypefn


@c finv scripts/statistics/distributions/finv.m
@anchor{XREFfinv}
@deftypefn {Function File} {} finv (@var{x}, @var{m}, @var{n})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the F distribution with @var{m} and @var{n} degrees of freedom.
@end deftypefn


@c gampdf scripts/statistics/distributions/gampdf.m
@anchor{XREFgampdf}
@deftypefn {Function File} {} gampdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, return the probability density function
(PDF) at @var{x} of the Gamma distribution with shape parameter @var{a} and
scale @var{b}.
@end deftypefn


@c gamcdf scripts/statistics/distributions/gamcdf.m
@anchor{XREFgamcdf}
@deftypefn {Function File} {} gamcdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the Gamma distribution with shape parameter @var{a} and
scale @var{b}.
@end deftypefn


@c gaminv scripts/statistics/distributions/gaminv.m
@anchor{XREFgaminv}
@deftypefn {Function File} {} gaminv (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the Gamma distribution with shape parameter @var{a} and
scale @var{b}.
@end deftypefn


@c geopdf scripts/statistics/distributions/geopdf.m
@anchor{XREFgeopdf}
@deftypefn {Function File} {} geopdf (@var{x}, @var{p})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the geometric distribution with parameter @var{p}.

The geometric distribution models the number of failures (@var{x}-1) of a
Bernoulli trial with probability @var{p} before the first success (@var{x}).
@end deftypefn


@c geocdf scripts/statistics/distributions/geocdf.m
@anchor{XREFgeocdf}
@deftypefn {Function File} {} geocdf (@var{x}, @var{p})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the geometric distribution with parameter @var{p}.

The geometric distribution models the number of failures (@var{x}-1) of a
Bernoulli trial with probability @var{p} before the first success (@var{x}).
@end deftypefn


@c geoinv scripts/statistics/distributions/geoinv.m
@anchor{XREFgeoinv}
@deftypefn {Function File} {} geoinv (@var{x}, @var{p})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the geometric distribution with parameter @var{p}.

The geometric distribution models the number of failures (@var{x}-1) of a
Bernoulli trial with probability @var{p} before the first success (@var{x}).
@end deftypefn


@c hygepdf scripts/statistics/distributions/hygepdf.m
@anchor{XREFhygepdf}
@deftypefn {Function File} {} hygepdf (@var{x}, @var{t}, @var{m}, @var{n})
Compute the probability density function (PDF) at @var{x} of the
hypergeometric distribution with parameters @var{t}, @var{m}, and @var{n}.

This is the probability of obtaining @var{x} marked items when randomly
drawing a sample of size @var{n} without replacement from a population of
total size @var{t} containing @var{m} marked items.

The parameters @var{t}, @var{m}, and @var{n} must be positive integers
with @var{m} and @var{n} not greater than @var{t}.
@end deftypefn


@c hygecdf scripts/statistics/distributions/hygecdf.m
@anchor{XREFhygecdf}
@deftypefn {Function File} {} hygecdf (@var{x}, @var{t}, @var{m}, @var{n})
Compute the cumulative distribution function (CDF) at @var{x} of the
hypergeometric distribution with parameters @var{t}, @var{m}, and @var{n}.

This is the probability of obtaining not more than @var{x} marked items
when randomly drawing a sample of size @var{n} without replacement from a
population of total size @var{t} containing @var{m} marked items.

The parameters @var{t}, @var{m}, and @var{n} must be positive integers
with @var{m} and @var{n} not greater than @var{t}.
@end deftypefn


@c hygeinv scripts/statistics/distributions/hygeinv.m
@anchor{XREFhygeinv}
@deftypefn {Function File} {} hygeinv (@var{x}, @var{t}, @var{m}, @var{n})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the hypergeometric distribution with parameters
@var{t}, @var{m}, and @var{n}.

This is the probability of obtaining @var{x} marked items when randomly
drawing a sample of size @var{n} without replacement from a population of
total size @var{t} containing @var{m} marked items.

The parameters @var{t}, @var{m}, and @var{n} must be positive integers
with @var{m} and @var{n} not greater than @var{t}.
@end deftypefn


@c kolmogorov_smirnov_cdf scripts/statistics/distributions/kolmogorov_smirnov_cdf.m
@anchor{XREFkolmogorov_smirnov_cdf}
@deftypefn {Function File} {} kolmogorov_smirnov_cdf (@var{x}, @var{tol})
Return the cumulative distribution function (CDF) at @var{x} of the
Kolmogorov-Smirnov distribution.

This is defined as
@tex
$$ Q(x) = \sum_{k=-\infty}^\infty (-1)^k \exp (-2 k^2 x^2) $$
@end tex
@ifnottex

@example
@group
         Inf
Q(x) =   SUM    (-1)^k exp (-2 k^2 x^2)
       k = -Inf
@end group
@end example

@end ifnottex
@noindent
for @var{x} > 0.

The optional parameter @var{tol} specifies the precision up to which
the series should be evaluated; the default is @var{tol} = @code{eps}.
@end deftypefn


@c laplace_pdf scripts/statistics/distributions/laplace_pdf.m
@anchor{XREFlaplace_pdf}
@deftypefn {Function File} {} laplace_pdf (@var{x})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the Laplace distribution.
@end deftypefn


@c laplace_cdf scripts/statistics/distributions/laplace_cdf.m
@anchor{XREFlaplace_cdf}
@deftypefn {Function File} {} laplace_cdf (@var{x})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the Laplace distribution.
@end deftypefn


@c laplace_inv scripts/statistics/distributions/laplace_inv.m
@anchor{XREFlaplace_inv}
@deftypefn {Function File} {} laplace_inv (@var{x})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the Laplace distribution.
@end deftypefn


@c logistic_pdf scripts/statistics/distributions/logistic_pdf.m
@anchor{XREFlogistic_pdf}
@deftypefn {Function File} {} logistic_pdf (@var{x})
For each element of @var{x}, compute the PDF at @var{x} of the
logistic distribution.
@end deftypefn


@c logistic_cdf scripts/statistics/distributions/logistic_cdf.m
@anchor{XREFlogistic_cdf}
@deftypefn {Function File} {} logistic_cdf (@var{x})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the logistic distribution.
@end deftypefn


@c logistic_inv scripts/statistics/distributions/logistic_inv.m
@anchor{XREFlogistic_inv}
@deftypefn {Function File} {} logistic_inv (@var{x})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the logistic distribution.
@end deftypefn


@c lognpdf scripts/statistics/distributions/lognpdf.m
@anchor{XREFlognpdf}
@deftypefn  {Function File} {} lognpdf (@var{x})
@deftypefnx {Function File} {} lognpdf (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the lognormal distribution with parameters
@var{mu} and @var{sigma}.

If a random variable follows this distribution, its logarithm is normally
distributed with mean @var{mu} and standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c logncdf scripts/statistics/distributions/logncdf.m
@anchor{XREFlogncdf}
@deftypefn  {Function File} {} logncdf (@var{x})
@deftypefnx {Function File} {} logncdf (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the lognormal distribution with parameters
@var{mu} and @var{sigma}.

If a random variable follows this distribution, its logarithm is normally
distributed with mean @var{mu} and standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c logninv scripts/statistics/distributions/logninv.m
@anchor{XREFlogninv}
@deftypefn  {Function File} {} logninv (@var{x})
@deftypefnx {Function File} {} logninv (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the lognormal distribution with parameters
@var{mu} and @var{sigma}.

If a random variable follows this distribution, its logarithm is normally
distributed with mean @var{mu} and standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c nbinpdf scripts/statistics/distributions/nbinpdf.m
@anchor{XREFnbinpdf}
@deftypefn {Function File} {} nbinpdf (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the negative binomial distribution with parameters
@var{n} and @var{p}.

When @var{n} is integer this is the Pascal distribution.
When @var{n} is extended to real numbers this is the Polya distribution.

The number of failures in a Bernoulli experiment with success probability
@var{p} before the @var{n}-th success follows this distribution.
@end deftypefn


@c nbincdf scripts/statistics/distributions/nbincdf.m
@anchor{XREFnbincdf}
@deftypefn {Function File} {} nbincdf (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the negative binomial distribution with parameters
@var{n} and @var{p}.

When @var{n} is integer this is the Pascal distribution.
When @var{n} is extended to real numbers this is the Polya distribution.

The number of failures in a Bernoulli experiment with success probability
@var{p} before the @var{n}-th success follows this distribution.
@end deftypefn


@c nbininv scripts/statistics/distributions/nbininv.m
@anchor{XREFnbininv}
@deftypefn {Function File} {} nbininv (@var{x}, @var{n}, @var{p})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the negative binomial distribution with parameters
@var{n} and @var{p}.

When @var{n} is integer this is the Pascal distribution.
When @var{n} is extended to real numbers this is the Polya distribution.

The number of failures in a Bernoulli experiment with success probability
@var{p} before the @var{n}-th success follows this distribution.
@end deftypefn


@c normpdf scripts/statistics/distributions/normpdf.m
@anchor{XREFnormpdf}
@deftypefn  {Function File} {} normpdf (@var{x})
@deftypefnx {Function File} {} normpdf (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the normal distribution with mean @var{mu} and
standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c normcdf scripts/statistics/distributions/normcdf.m
@anchor{XREFnormcdf}
@deftypefn  {Function File} {} normcdf (@var{x})
@deftypefnx {Function File} {} normcdf (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the normal distribution with mean @var{mu} and
standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c norminv scripts/statistics/distributions/norminv.m
@anchor{XREFnorminv}
@deftypefn  {Function File} {} norminv (@var{x})
@deftypefnx {Function File} {} norminv (@var{x}, @var{mu}, @var{sigma})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the normal distribution with mean @var{mu} and
standard deviation @var{sigma}.

Default values are @var{mu} = 0, @var{sigma} = 1.
@end deftypefn


@c poisspdf scripts/statistics/distributions/poisspdf.m
@anchor{XREFpoisspdf}
@deftypefn {Function File} {} poisspdf (@var{x}, @var{lambda})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the Poisson distribution with parameter @var{lambda}.
@end deftypefn


@c poisscdf scripts/statistics/distributions/poisscdf.m
@anchor{XREFpoisscdf}
@deftypefn {Function File} {} poisscdf (@var{x}, @var{lambda})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the Poisson distribution with parameter @var{lambda}.
@end deftypefn


@c poissinv scripts/statistics/distributions/poissinv.m
@anchor{XREFpoissinv}
@deftypefn {Function File} {} poissinv (@var{x}, @var{lambda})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the Poisson distribution with parameter @var{lambda}.
@end deftypefn


@c stdnormal_pdf scripts/statistics/distributions/stdnormal_pdf.m
@anchor{XREFstdnormal_pdf}
@deftypefn {Function File} {} stdnormal_pdf (@var{x})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the standard normal distribution
(mean = 0, standard deviation = 1).
@end deftypefn


@c stdnormal_cdf scripts/statistics/distributions/stdnormal_cdf.m
@anchor{XREFstdnormal_cdf}
@deftypefn {Function File} {} stdnormal_cdf (@var{x})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the standard normal distribution
(mean = 0, standard deviation = 1).
@end deftypefn


@c stdnormal_inv scripts/statistics/distributions/stdnormal_inv.m
@anchor{XREFstdnormal_inv}
@deftypefn {Function File} {} stdnormal_inv (@var{x})
For each element of @var{x}, compute the quantile (the
inverse of the CDF) at @var{x} of the standard normal distribution
(mean = 0, standard deviation = 1).
@end deftypefn


@c tpdf scripts/statistics/distributions/tpdf.m
@anchor{XREFtpdf}
@deftypefn {Function File} {} tpdf (@var{x}, @var{n})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the @var{t} (Student) distribution with
@var{n} degrees of freedom.
@end deftypefn


@c tcdf scripts/statistics/distributions/tcdf.m
@anchor{XREFtcdf}
@deftypefn {Function File} {} tcdf (@var{x}, @var{n})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the t (Student) distribution with
@var{n} degrees of freedom.
@end deftypefn


@c tinv scripts/statistics/distributions/tinv.m
@anchor{XREFtinv}
@deftypefn {Function File} {} tinv (@var{x}, @var{n})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the t (Student) distribution with @var{n}
degrees of freedom.

This function is analogous to looking in a table for the t-value of a
single-tailed distribution.
@end deftypefn


@c unidpdf scripts/statistics/distributions/unidpdf.m
@anchor{XREFunidpdf}
@deftypefn {Function File} {} unidpdf (@var{x}, @var{n})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of a discrete uniform distribution which assumes
the integer values 1--@var{n} with equal probability.

Warning: The underlying implementation uses the double class and will only
be accurate for @var{n} @leq{} @code{bitmax} (@w{@math{2^{53} - 1}} on
IEEE 754 compatible systems).
@end deftypefn


@c unidcdf scripts/statistics/distributions/unidcdf.m
@anchor{XREFunidcdf}
@deftypefn {Function File} {} unidcdf (@var{x}, @var{n})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of a discrete uniform distribution which assumes
the integer values 1--@var{n} with equal probability.
@end deftypefn


@c unidinv scripts/statistics/distributions/unidinv.m
@anchor{XREFunidinv}
@deftypefn {Function File} {} unidinv (@var{x}, @var{n})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the discrete uniform distribution which assumes
the integer values 1--@var{n} with equal probability.
@end deftypefn


@c unifpdf scripts/statistics/distributions/unifpdf.m
@anchor{XREFunifpdf}
@deftypefn  {Function File} {} unifpdf (@var{x})
@deftypefnx {Function File} {} unifpdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the probability density function (PDF)
at @var{x} of the uniform distribution on the interval [@var{a}, @var{b}].

Default values are @var{a} = 0, @var{b} = 1.
@end deftypefn


@c unifcdf scripts/statistics/distributions/unifcdf.m
@anchor{XREFunifcdf}
@deftypefn  {Function File} {} unifcdf (@var{x})
@deftypefnx {Function File} {} unifcdf (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the cumulative distribution function
(CDF) at @var{x} of the uniform distribution on the interval
[@var{a}, @var{b}].

Default values are @var{a} = 0, @var{b} = 1.
@end deftypefn


@c unifinv scripts/statistics/distributions/unifinv.m
@anchor{XREFunifinv}
@deftypefn  {Function File} {} unifinv (@var{x})
@deftypefnx {Function File} {} unifinv (@var{x}, @var{a}, @var{b})
For each element of @var{x}, compute the quantile (the inverse of the CDF)
at @var{x} of the uniform distribution on the interval [@var{a}, @var{b}].

Default values are @var{a} = 0, @var{b} = 1.
@end deftypefn


@c wblpdf scripts/statistics/distributions/wblpdf.m
@anchor{XREFwblpdf}
@deftypefn  {Function File} {} wblpdf (@var{x})
@deftypefnx {Function File} {} wblpdf (@var{x}, @var{scale})
@deftypefnx {Function File} {} wblpdf (@var{x}, @var{scale}, @var{shape})
Compute the probability density function (PDF) at @var{x} of the
Weibull distribution with scale parameter @var{scale} and
shape parameter @var{shape}.

This is given by
@tex
$$  {shape \over scale^{shape}} \cdot x^{shape-1} \cdot e^{-({x \over scale})^{shape}} $$
@end tex
@ifnottex

@example
shape * scale^(-shape) * x^(shape-1) * exp (-(x/scale)^shape)
@end example

@end ifnottex
@noindent
for @var{x} @geq{} 0.

Default values are @var{scale} = 1, @var{shape} = 1.
@end deftypefn


@c wblcdf scripts/statistics/distributions/wblcdf.m
@anchor{XREFwblcdf}
@deftypefn  {Function File} {} wblcdf (@var{x})
@deftypefnx {Function File} {} wblcdf (@var{x}, @var{scale})
@deftypefnx {Function File} {} wblcdf (@var{x}, @var{scale}, @var{shape})
Compute the cumulative distribution function (CDF) at @var{x} of the
Weibull distribution with scale parameter @var{scale} and shape
parameter @var{shape}.

This is defined as
@tex
$$ 1 - e^{-({x \over scale})^{shape}} $$
for $x \geq 0$.
@end tex
@ifnottex

@example
1 - exp (-(x/scale)^shape)
@end example

@noindent
for @var{x} @geq{} 0.

Default values are @var{scale} = 1, @var{shape} = 1.
@end ifnottex
@end deftypefn


@c wblinv scripts/statistics/distributions/wblinv.m
@anchor{XREFwblinv}
@deftypefn  {Function File} {} wblinv (@var{x})
@deftypefnx {Function File} {} wblinv (@var{x}, @var{scale})
@deftypefnx {Function File} {} wblinv (@var{x}, @var{scale}, @var{shape})
Compute the quantile (the inverse of the CDF) at @var{x} of the
Weibull distribution with scale parameter @var{scale} and
shape parameter @var{shape}.

Default values are @var{scale} = 1, @var{shape} = 1.
@end deftypefn


@node Tests
@section Tests

Octave can perform many different statistical tests.  The following
table summarizes the available tests.

@tex
\vskip 6pt
{\hbox to \hsize {\hfill\vbox{\offinterlineskip \tabskip=0pt
\halign{
\vrule height2.0ex depth1.ex width 0.6pt #\tabskip=0.3em &
# \hfil & \vrule # & # \hfil & # \vrule width 0.6pt \tabskip=0pt\cr
\noalign{\hrule height 0.6pt}
& @strong{Hypothesis} && {\bf Test Functions} &\cr
\noalign{\hrule}
& Equal mean values && anova, hotelling\_test2, t\_test\_2, &\cr
&                   && welch\_test, wilcoxon\_test, z\_test\_2 &\cr
& Equal medians && kruskal\_wallis\_test, sign\_test &\cr
& Equal variances && bartlett\_test, manova, var\_test &\cr
& Equal distributions && chisquare\_test\_homogeneity, &\cr
&                     && kolmogorov\_smirnov\_test\_2, u\_test &\cr
& Equal marginal frequencies && mcnemar\_test &\cr
& Equal success probabilities && prop\_test\_2 &\cr
& Independent observations && chisquare\_test\_independence, &\cr
&                          && run\_test &\cr
& Uncorrelated observations && cor\_test &\cr
& Given mean value && hotelling\_test, t\_test, z\_test &\cr
& Observations from distribution && kolmogorov\_smirnov\_test &\cr
& Regression && f\_test\_regression, t\_test\_regression &\cr
\noalign{\hrule height 0.6pt}
}}\hfill}}
@end tex
@ifnottex
@multitable @columnfractions .4 .5
@headitem Hypothesis
  @tab Test Functions
@item Equal mean values
  @tab @code{anova}, @code{hotelling_test2}, @code{t_test_2},
       @code{welch_test}, @code{wilcoxon_test}, @code{z_test_2}
@item Equal medians
  @tab @code{kruskal_wallis_test}, @code{sign_test}
@item Equal variances
  @tab @code{bartlett_test}, @code{manova}, @code{var_test}
@item Equal distributions
  @tab @code{chisquare_test_homogeneity}, @code{kolmogorov_smirnov_test_2},
       @code{u_test}
@item Equal marginal frequencies
  @tab @code{mcnemar_test}
@item Equal success probabilities
  @tab @code{prop_test_2}
@item Independent observations
  @tab @code{chisquare_test_independence}, @code{run_test}
@item Uncorrelated observations
  @tab @code{cor_test}
@item Given mean value
  @tab @code{hotelling_test}, @code{t_test}, @code{z_test}
@item Observations from given distribution
  @tab @code{kolmogorov_smirnov_test}
@item Regression
  @tab @code{f_test_regression}, @code{t_test_regression}
@end multitable
@end ifnottex

The tests return a p-value that describes the outcome of the test.
Assuming that the test hypothesis is true, the p-value is the probability
of obtaining a worse result than the observed one.  So large p-values
corresponds to a successful test.  Usually a test hypothesis is accepted
if the p-value exceeds 0.05.

@c anova scripts/statistics/tests/anova.m
@anchor{XREFanova}
@deftypefn {Function File} {[@var{pval}, @var{f}, @var{df_b}, @var{df_w}] =} anova (@var{y}, @var{g})
Perform a one-way analysis of variance (ANOVA).

The goal is to test whether the population means of data taken from
@var{k} different groups are all equal.

Data may be given in a single vector @var{y} with groups specified by a
corresponding vector of group labels @var{g} (e.g., numbers from 1 to
@var{k}).  This is the general form which does not impose any restriction
on the number of data in each group or the group labels.

If @var{y} is a matrix and @var{g} is omitted, each column of @var{y} is
treated as a group.  This form is only appropriate for balanced ANOVA in
which the numbers of samples from each group are all equal.

Under the null of constant means, the statistic @var{f} follows an F
distribution with @var{df_b} and @var{df_w} degrees of freedom.

The p-value (1 minus the CDF of this distribution at @var{f}) is returned
in @var{pval}.

If no output argument is given, the standard one-way ANOVA table is printed.
@seealso{@ref{XREFmanova,,manova}}
@end deftypefn


@c bartlett_test scripts/statistics/tests/bartlett_test.m
@anchor{XREFbartlett_test}
@deftypefn {Function File} {[@var{pval}, @var{chisq}, @var{df}] =} bartlett_test (@var{x1}, @dots{})
Perform a Bartlett test for the homogeneity of variances in the data
vectors @var{x1}, @var{x2}, @dots{}, @var{xk}, where @var{k} > 1.

Under the null of equal variances, the test statistic @var{chisq}
approximately follows a chi-square distribution with @var{df} degrees of
freedom.

The p-value (1 minus the CDF of this distribution at @var{chisq}) is
returned in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c chisquare_test_homogeneity scripts/statistics/tests/chisquare_test_homogeneity.m
@anchor{XREFchisquare_test_homogeneity}
@deftypefn {Function File} {[@var{pval}, @var{chisq}, @var{df}] =} chisquare_test_homogeneity (@var{x}, @var{y}, @var{c})
Given two samples @var{x} and @var{y}, perform a chisquare test for
homogeneity of the null hypothesis that @var{x} and @var{y} come from
the same distribution, based on the partition induced by the
(strictly increasing) entries of @var{c}.

For large samples, the test statistic @var{chisq} approximately follows a
chisquare distribution with @var{df} = @code{length (@var{c})} degrees of
freedom.

The p-value (1 minus the CDF of this distribution at @var{chisq}) is
returned in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c chisquare_test_independence scripts/statistics/tests/chisquare_test_independence.m
@anchor{XREFchisquare_test_independence}
@deftypefn {Function File} {[@var{pval}, @var{chisq}, @var{df}] =} chisquare_test_independence (@var{x})
Perform a chi-square test for independence based on the contingency table
@var{x}.

Under the null hypothesis of independence, @var{chisq} approximately has a
chi-square distribution with @var{df} degrees of freedom.

The p-value (1 minus the CDF of this distribution at chisq) of the test is
returned in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c cor_test scripts/statistics/tests/cor_test.m
@anchor{XREFcor_test}
@deftypefn {Function File} {} cor_test (@var{x}, @var{y}, @var{alt}, @var{method})
Test whether two samples @var{x} and @var{y} come from uncorrelated
populations.

The optional argument string @var{alt} describes the alternative
hypothesis, and can be @qcode{"!="} or @qcode{"<>"} (nonzero), @qcode{">"}
(greater than 0), or @qcode{"<"} (less than 0).  The default is the
two-sided case.

The optional argument string @var{method} specifies which correlation
coefficient to use for testing.  If @var{method} is @qcode{"pearson"}
(default), the (usual) Pearson's produt moment correlation coefficient is
used.  In this case, the data should come from a bivariate normal
distribution.  Otherwise, the other two methods offer nonparametric
alternatives.  If @var{method} is @qcode{"kendall"}, then Kendall's rank
correlation tau is used.  If @var{method} is @qcode{"spearman"}, then
Spearman's rank correlation rho is used.  Only the first character is
necessary.

The output is a structure with the following elements:

@table @var
@item pval
The p-value of the test.

@item stat
The value of the test statistic.

@item dist
The distribution of the test statistic.

@item params
The parameters of the null distribution of the test statistic.

@item alternative
The alternative hypothesis.

@item method
The method used for testing.
@end table

If no output argument is given, the p-value is displayed.
@end deftypefn


@c f_test_regression scripts/statistics/tests/f_test_regression.m
@anchor{XREFf_test_regression}
@deftypefn {Function File} {[@var{pval}, @var{f}, @var{df_num}, @var{df_den}] =} f_test_regression (@var{y}, @var{x}, @var{rr}, @var{r})
Perform an F test for the null hypothesis @nospell{rr * b = r} in a
classical normal regression model y = X * b + e.

Under the null, the test statistic @var{f} follows an F distribution with
@var{df_num} and @var{df_den} degrees of freedom.

The p-value (1 minus the CDF of this distribution at @var{f}) is returned
in @var{pval}.

If not given explicitly, @var{r} = 0.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c hotelling_test scripts/statistics/tests/hotelling_test.m
@anchor{XREFhotelling_test}
@deftypefn {Function File} {[@var{pval}, @var{tsq}] =} hotelling_test (@var{x}, @var{m})
For a sample @var{x} from a multivariate normal distribution with unknown
mean and covariance matrix, test the null hypothesis that
@code{mean (@var{x}) == @var{m}}.

Hotelling's @math{T^2} is returned in @var{tsq}.  Under the null,
@math{(n-p) T^2 / (p(n-1))} has an F distribution with @math{p} and
@math{n-p} degrees of freedom, where @math{n} and @math{p} are the
numbers of samples and variables, respectively.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c hotelling_test_2 scripts/statistics/tests/hotelling_test_2.m
@anchor{XREFhotelling_test_2}
@deftypefn {Function File} {[@var{pval}, @var{tsq}] =} hotelling_test_2 (@var{x}, @var{y})
For two samples @var{x} from multivariate normal distributions with
the same number of variables (columns), unknown means and unknown
equal covariance matrices, test the null hypothesis @code{mean
(@var{x}) == mean (@var{y})}.

Hotelling's two-sample @math{T^2} is returned in @var{tsq}.  Under the null,
@tex
$$
{(n_x+n_y-p-1) T^2 \over p(n_x+n_y-2)}
$$
@end tex
@ifnottex

@example
(n_x+n_y-p-1) T^2 / (p(n_x+n_y-2))
@end example

@end ifnottex
@noindent
has an F distribution with @math{p} and @math{n_x+n_y-p-1} degrees of
freedom, where @math{n_x} and @math{n_y} are the sample sizes and
@math{p} is the number of variables.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c kolmogorov_smirnov_test scripts/statistics/tests/kolmogorov_smirnov_test.m
@anchor{XREFkolmogorov_smirnov_test}
@deftypefn {Function File} {[@var{pval}, @var{ks}] =} kolmogorov_smirnov_test (@var{x}, @var{dist}, @var{params}, @var{alt})
Perform a Kolmogorov-Smirnov test of the null hypothesis that the
sample @var{x} comes from the (continuous) distribution @var{dist}.

if F and G are the CDFs corresponding to the sample and dist,
respectively, then the null is that F == G.

The optional argument @var{params} contains a list of parameters of
@var{dist}.  For example, to test whether a sample @var{x} comes from
a uniform distribution on [2,4], use

@example
kolmogorov_smirnov_test (x, "unif", 2, 4)
@end example

@noindent
@var{dist} can be any string for which a function @var{distcdf}
that calculates the CDF of distribution @var{dist} exists.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative F != G@.  In this case, the
test statistic @var{ks} follows a two-sided Kolmogorov-Smirnov
distribution.  If @var{alt} is @qcode{">"}, the one-sided alternative F >
G is considered.  Similarly for @qcode{"<"}, the one-sided alternative F >
G is considered.  In this case, the test statistic @var{ks} has a
one-sided Kolmogorov-Smirnov distribution.  The default is the two-sided
case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c kolmogorov_smirnov_test_2 scripts/statistics/tests/kolmogorov_smirnov_test_2.m
@anchor{XREFkolmogorov_smirnov_test_2}
@deftypefn {Function File} {[@var{pval}, @var{ks}, @var{d}] =} kolmogorov_smirnov_test_2 (@var{x}, @var{y}, @var{alt})
Perform a 2-sample Kolmogorov-Smirnov test of the null hypothesis that the
samples @var{x} and @var{y} come from the same (continuous) distribution.

If F and G are the CDFs corresponding to the @var{x} and @var{y} samples,
respectively, then the null is that F == G.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative F != G@.  In this case, the
test statistic @var{ks} follows a two-sided Kolmogorov-Smirnov
distribution.  If @var{alt} is @qcode{">"}, the one-sided alternative F >
G is considered.  Similarly for @qcode{"<"}, the one-sided alternative F <
G is considered.  In this case, the test statistic @var{ks} has a
one-sided Kolmogorov-Smirnov distribution.  The default is the two-sided
case.

The p-value of the test is returned in @var{pval}.

The third returned value, @var{d}, is the test statistic, the maximum
vertical distance between the two cumulative distribution functions.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c kruskal_wallis_test scripts/statistics/tests/kruskal_wallis_test.m
@anchor{XREFkruskal_wallis_test}
@deftypefn {Function File} {[@var{pval}, @var{k}, @var{df}] =} kruskal_wallis_test (@var{x1}, @dots{})
Perform a @nospell{Kruskal-Wallis} one-factor analysis of variance.

Suppose a variable is observed for @var{k} > 1 different groups, and let
@var{x1}, @dots{}, @var{xk} be the corresponding data vectors.

Under the null hypothesis that the ranks in the pooled sample are not
affected by the group memberships, the test statistic @var{k} is
approximately chi-square with @var{df} = @var{k} - 1 degrees of freedom.

If the data contains ties (some value appears more than once)
@var{k} is divided by

1 - @var{sum_ties} / (@var{n}^3 - @var{n})

where @var{sum_ties} is the sum of @var{t}^2 - @var{t} over each group of
ties where @var{t} is the number of ties in the group and @var{n} is the
total number of values in the input data.  For more info on this
adjustment see @nospell{William H. Kruskal and W. Allen Wallis},
@cite{Use of Ranks in One-Criterion Variance Analysis},
Journal of the American Statistical Association, Vol. 47, No. 260 (Dec 1952).

The p-value (1 minus the CDF of this distribution at @var{k}) is returned
in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c manova scripts/statistics/tests/manova.m
@anchor{XREFmanova}
@deftypefn {Function File} {} manova (@var{x}, @var{g})
Perform a one-way multivariate analysis of variance (MANOVA).

The goal is to test whether the p-dimensional population means of data
taken from @var{k} different groups are all equal.  All data are assumed
drawn independently from p-dimensional normal distributions with the same
covariance matrix.

The data matrix is given by @var{x}.  As usual, rows are observations and
columns are variables.  The vector @var{g} specifies the corresponding
group labels (e.g., numbers from 1 to @var{k}).

The LR test statistic (@nospell{Wilks' Lambda}) and approximate p-values are
computed and displayed.
@seealso{@ref{XREFanova,,anova}}
@end deftypefn


@c mcnemar_test scripts/statistics/tests/mcnemar_test.m
@anchor{XREFmcnemar_test}
@deftypefn {Function File} {[@var{pval}, @var{chisq}, @var{df}] =} mcnemar_test (@var{x})
For a square contingency table @var{x} of data cross-classified on the row
and column variables, @nospell{McNemar's} test can be used for testing the
null hypothesis of symmetry of the classification probabilities.

Under the null, @var{chisq} is approximately distributed as chisquare with
@var{df} degrees of freedom.

The p-value (1 minus the CDF of this distribution at @var{chisq}) is
returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c prop_test_2 scripts/statistics/tests/prop_test_2.m
@anchor{XREFprop_test_2}
@deftypefn {Function File} {[@var{pval}, @var{z}] =} prop_test_2 (@var{x1}, @var{n1}, @var{x2}, @var{n2}, @var{alt})
If @var{x1} and @var{n1} are the counts of successes and trials in one
sample, and @var{x2} and @var{n2} those in a second one, test the null
hypothesis that the success probabilities @var{p1} and @var{p2} are the
same.

Under the null, the test statistic @var{z} approximately follows a
standard normal distribution.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative @var{p1} != @var{p2}.  If
@var{alt} is @qcode{">"}, the one-sided alternative @var{p1} > @var{p2} is
used.  Similarly for @qcode{"<"}, the one-sided alternative
@var{p1} < @var{p2} is used.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c run_test scripts/statistics/tests/run_test.m
@anchor{XREFrun_test}
@deftypefn {Function File} {[@var{pval}, @var{chisq}] =} run_test (@var{x})
Perform a chi-square test with 6 degrees of freedom based on the upward
runs in the columns of @var{x}.

@code{run_test} can be used to decide whether @var{x} contains independent
data.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value is displayed.
@end deftypefn


@c sign_test scripts/statistics/tests/sign_test.m
@anchor{XREFsign_test}
@deftypefn {Function File} {[@var{pval}, @var{b}, @var{n}] =} sign_test (@var{x}, @var{y}, @var{alt})
For two matched-pair samples @var{x} and @var{y}, perform a sign test
of the null hypothesis
PROB (@var{x} > @var{y}) == PROB (@var{x} < @var{y}) == 1/2.

Under the null, the test statistic @var{b} roughly follows a
binomial distribution with parameters
@code{@var{n} = sum (@var{x} != @var{y})} and @var{p} = 1/2.

With the optional argument @code{alt}, the alternative of interest can be
selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
hypothesis is tested against the two-sided alternative
PROB (@var{x} < @var{y}) != 1/2.  If @var{alt} is @qcode{">"}, the one-sided
alternative PROB (@var{x} > @var{y}) > 1/2 ("x is stochastically greater
than y") is considered.  Similarly for @qcode{"<"}, the one-sided
alternative PROB (@var{x} > @var{y}) < 1/2 ("x is stochastically less than
y") is considered.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c t_test scripts/statistics/tests/t_test.m
@anchor{XREFt_test}
@deftypefn {Function File} {[@var{pval}, @var{t}, @var{df}] =} t_test (@var{x}, @var{m}, @var{alt})
For a sample @var{x} from a normal distribution with unknown mean and
variance, perform a t-test of the null hypothesis
@code{mean (@var{x}) == @var{m}}.

Under the null, the test statistic @var{t} follows a Student distribution
with @code{@var{df} = length (@var{x}) - 1} degrees of freedom.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative @code{mean (@var{x}) !=
@var{m}}.  If @var{alt} is @qcode{">"}, the one-sided alternative
@code{mean (@var{x}) > @var{m}} is considered.  Similarly for @var{"<"},
the one-sided alternative @code{mean (@var{x}) < @var{m}} is considered.
The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c t_test_2 scripts/statistics/tests/t_test_2.m
@anchor{XREFt_test_2}
@deftypefn {Function File} {[@var{pval}, @var{t}, @var{df}] =} t_test_2 (@var{x}, @var{y}, @var{alt})
For two samples x and y from normal distributions with unknown means and
unknown equal variances, perform a two-sample t-test of the null
hypothesis of equal means.

Under the null, the test statistic @var{t} follows a Student distribution
with @var{df} degrees of freedom.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative @code{mean (@var{x}) != mean
(@var{y})}.  If @var{alt} is @qcode{">"}, the one-sided alternative
@code{mean (@var{x}) > mean (@var{y})} is used.  Similarly for
@qcode{"<"}, the one-sided alternative @code{mean (@var{x}) < mean
(@var{y})} is used.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c t_test_regression scripts/statistics/tests/t_test_regression.m
@anchor{XREFt_test_regression}
@deftypefn {Function File} {[@var{pval}, @var{t}, @var{df}] =} t_test_regression (@var{y}, @var{x}, @var{rr}, @var{r}, @var{alt})
Perform a t test for the null hypothesis
@nospell{@code{@var{rr} * @var{b} = @var{r}}} in a classical normal
regression model @code{@var{y} = @var{x} * @var{b} + @var{e}}.

Under the null, the test statistic @var{t} follows a @var{t} distribution
with @var{df} degrees of freedom.

If @var{r} is omitted, a value of 0 is assumed.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative @nospell{@code{@var{rr} *
@var{b} != @var{r}}}.  If @var{alt} is @qcode{">"}, the one-sided
alternative @nospell{@code{@var{rr} * @var{b} > @var{r}}} is used.
Similarly for @var{"<"}, the one-sided alternative @nospell{@code{@var{rr}
* @var{b} < @var{r}}} is used.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c u_test scripts/statistics/tests/u_test.m
@anchor{XREFu_test}
@deftypefn {Function File} {[@var{pval}, @var{z}] =} u_test (@var{x}, @var{y}, @var{alt})
For two samples @var{x} and @var{y}, perform a Mann-Whitney U-test of
the null hypothesis
PROB (@var{x} > @var{y}) == 1/2 == PROB (@var{x} < @var{y}).

Under the null, the test statistic @var{z} approximately follows a
standard normal distribution.  Note that this test is equivalent to the
Wilcoxon rank-sum test.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative
PROB (@var{x} > @var{y}) != 1/2.  If @var{alt} is @qcode{">"}, the one-sided
alternative PROB (@var{x} > @var{y}) > 1/2 is considered.  Similarly for
@qcode{"<"}, the one-sided alternative PROB (@var{x} > @var{y}) < 1/2 is
considered.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c var_test scripts/statistics/tests/var_test.m
@anchor{XREFvar_test}
@deftypefn {Function File} {[@var{pval}, @var{f}, @var{df_num}, @var{df_den}] =} var_test (@var{x}, @var{y}, @var{alt})
For two samples @var{x} and @var{y} from normal distributions with
unknown means and unknown variances, perform an F-test of the null
hypothesis of equal variances.

Under the null, the test statistic @var{f} follows an F-distribution with
@var{df_num} and @var{df_den} degrees of freedom.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative @code{var (@var{x}) != var
(@var{y})}.  If @var{alt} is @qcode{">"}, the one-sided alternative
@code{var (@var{x}) > var (@var{y})} is used.  Similarly for "<", the
one-sided alternative @code{var (@var{x}) > var (@var{y})} is used.  The
default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c welch_test scripts/statistics/tests/welch_test.m
@anchor{XREFwelch_test}
@deftypefn {Function File} {[@var{pval}, @var{t}, @var{df}] =} welch_test (@var{x}, @var{y}, @var{alt})
For two samples @var{x} and @var{y} from normal distributions with
unknown means and unknown and not necessarily equal variances,
perform a Welch test of the null hypothesis of equal means.

Under the null, the test statistic @var{t} approximately follows a
Student distribution with @var{df} degrees of freedom.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative
@code{mean (@var{x}) != @var{m}}.  If @var{alt} is @qcode{">"}, the
one-sided alternative mean(x) > @var{m} is considered.  Similarly for
@qcode{"<"}, the one-sided alternative mean(x) < @var{m} is considered.
The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c wilcoxon_test scripts/statistics/tests/wilcoxon_test.m
@anchor{XREFwilcoxon_test}
@deftypefn {Function File} {[@var{pval}, @var{z}] =} wilcoxon_test (@var{x}, @var{y}, @var{alt})
For two matched-pair sample vectors @var{x} and @var{y}, perform a
Wilcoxon signed-rank test of the null hypothesis
PROB (@var{x} > @var{y}) == 1/2.

Under the null, the test statistic @var{z} approximately follows a
standard normal distribution when @var{n} > 25.

@strong{Caution:} This function assumes a normal distribution for @var{z}
and thus is invalid for @var{n} @leq{} 25.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative
PROB (@var{x} > @var{y}) != 1/2.  If alt is @qcode{">"}, the one-sided
alternative PROB (@var{x} > @var{y}) > 1/2 is considered.  Similarly for
@qcode{"<"}, the one-sided alternative PROB (@var{x} > @var{y}) < 1/2 is
considered.  The default is the two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed.
@end deftypefn


@c z_test scripts/statistics/tests/z_test.m
@anchor{XREFz_test}
@deftypefn {Function File} {[@var{pval}, @var{z}] =} z_test (@var{x}, @var{m}, @var{v}, @var{alt})
Perform a Z-test of the null hypothesis @code{mean (@var{x}) == @var{m}}
for a sample @var{x} from a normal distribution with unknown mean and known
variance @var{v}.

Under the null, the test statistic @var{z} follows a standard normal
distribution.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative
@code{mean (@var{x}) != @var{m}}.  If @var{alt} is @qcode{">"}, the
one-sided alternative @code{mean (@var{x}) > @var{m}} is considered.
Similarly for @qcode{"<"}, the one-sided alternative
@code{mean (@var{x}) < @var{m}} is considered.  The default is the two-sided
case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed along
with some information.
@end deftypefn


@c z_test_2 scripts/statistics/tests/z_test_2.m
@anchor{XREFz_test_2}
@deftypefn {Function File} {[@var{pval}, @var{z}] =} z_test_2 (@var{x}, @var{y}, @var{v_x}, @var{v_y}, @var{alt})
For two samples @var{x} and @var{y} from normal distributions with unknown
means and known variances @var{v_x} and @var{v_y}, perform a Z-test of the
hypothesis of equal means.

Under the null, the test statistic @var{z} follows a standard normal
distribution.

With the optional argument string @var{alt}, the alternative of interest
can be selected.  If @var{alt} is @qcode{"!="} or @qcode{"<>"}, the null
is tested against the two-sided alternative
@code{mean (@var{x}) != mean (@var{y})}.  If alt is @qcode{">"}, the
one-sided alternative @code{mean (@var{x}) > mean (@var{y})} is used.
Similarly for @qcode{"<"}, the one-sided alternative
@code{mean (@var{x}) < mean (@var{y})} is used.  The default is the
two-sided case.

The p-value of the test is returned in @var{pval}.

If no output argument is given, the p-value of the test is displayed along
with some information.
@end deftypefn


@node Random Number Generation
@section Random Number Generation

Octave can generate random numbers from a large number of distributions.
The random number generators are based on the random number generators
described in @ref{Special Utility Matrices}.
@c Should rand, randn, rande, randp, and randg be moved to here?

The following table summarizes the available random number generators
(in alphabetical order).

@tex
\vskip 6pt
{\hbox to \hsize {\hfill\vbox{\offinterlineskip \tabskip=0pt
\halign{
\vrule height2.0ex depth1.ex width 0.6pt #\tabskip=0.3em &
# \hfil & \vrule # & # \hfil & # \vrule width 0.6pt \tabskip=0pt\cr
\noalign{\hrule height 0.6pt}
& {\bf Distribution}                && {\bf Function} &\cr
\noalign{\hrule}
& Beta Distribution                 && betarnd &\cr
& Binomial Distribution             && binornd &\cr
& Cauchy Distribution               && cauchy\_rnd &\cr
& Chi-Square Distribution           && chi2rnd &\cr
& Univariate Discrete Distribution  && discrete\_rnd &\cr
& Empirical Distribution            && empirical\_rnd &\cr
& Exponential Distribution          && exprnd &\cr
& F Distribution                    && frnd &\cr
& Gamma Distribution                && gamrnd &\cr
& Geometric Distribution            && geornd &\cr
& Hypergeometric Distribution       && hygernd &\cr
& Laplace Distribution              && laplace\_rnd &\cr
& Logistic Distribution             && logistic\_rnd &\cr
& Log-Normal Distribution           && lognrnd &\cr
& Pascal Distribution               && nbinrnd &\cr
& Univariate Normal Distribution    && normrnd &\cr
& Poisson Distribution              && poissrnd &\cr
& Standard Normal Distribution      && stdnormal\_rnd &\cr
& t (Student) Distribution          && trnd &\cr
& Univariate Discrete Distribution  && unidrnd &\cr
& Uniform Distribution              && unifrnd &\cr
& Weibull Distribution              && wblrnd &\cr
& Wiener Process                    && wienrnd &\cr
\noalign{\hrule height 0.6pt}
}}\hfill}}
@end tex
@ifnottex
@multitable @columnfractions .4 .3
@headitem Distribution                  @tab Function
@item Beta Distribution                 @tab @code{betarnd}
@item Binomial Distribution             @tab @code{binornd}
@item Cauchy Distribution               @tab @code{cauchy_rnd}
@item Chi-Square Distribution           @tab @code{chi2rnd}
@item Univariate Discrete Distribution  @tab @code{discrete_rnd}
@item Empirical Distribution            @tab @code{empirical_rnd}
@item Exponential Distribution          @tab @code{exprnd}
@item F Distribution                    @tab @code{frnd}
@item Gamma Distribution                @tab @code{gamrnd}
@item Geometric Distribution            @tab @code{geornd}
@item Hypergeometric Distribution       @tab @code{hygernd}
@item Laplace Distribution              @tab @code{laplace_rnd}
@item Logistic Distribution             @tab @code{logistic_rnd}
@item Log-Normal Distribution           @tab @code{lognrnd}
@item Pascal Distribution               @tab @code{nbinrnd}
@item Univariate Normal Distribution    @tab @code{normrnd}
@item Poisson Distribution              @tab @code{poissrnd}
@item Standard Normal Distribution      @tab @code{stdnormal_rnd}
@item t (Student) Distribution          @tab @code{trnd}
@item Univariate Discrete Distribution  @tab @code{unidrnd}
@item Uniform Distribution              @tab @code{unifrnd}
@item Weibull Distribution              @tab @code{wblrnd}
@item Wiener Process                    @tab @code{wienrnd}
@end multitable
@end ifnottex

@c betarnd scripts/statistics/distributions/betarnd.m
@anchor{XREFbetarnd}
@deftypefn  {Function File} {} betarnd (@var{a}, @var{b})
@deftypefnx {Function File} {} betarnd (@var{a}, @var{b}, @var{r})
@deftypefnx {Function File} {} betarnd (@var{a}, @var{b}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} betarnd (@var{a}, @var{b}, [@var{sz}])
Return a matrix of random samples from the Beta distribution with parameters
@var{a} and @var{b}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{a} and @var{b}.
@end deftypefn


@c binornd scripts/statistics/distributions/binornd.m
@anchor{XREFbinornd}
@deftypefn  {Function File} {} binornd (@var{n}, @var{p})
@deftypefnx {Function File} {} binornd (@var{n}, @var{p}, @var{r})
@deftypefnx {Function File} {} binornd (@var{n}, @var{p}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} binornd (@var{n}, @var{p}, [@var{sz}])
Return a matrix of random samples from the binomial distribution with
parameters @var{n} and @var{p}, where @var{n} is the number of trials
and @var{p} is the probability of success.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{n} and @var{p}.
@end deftypefn


@c cauchy_rnd scripts/statistics/distributions/cauchy_rnd.m
@anchor{XREFcauchy_rnd}
@deftypefn  {Function File} {} cauchy_rnd (@var{location}, @var{scale})
@deftypefnx {Function File} {} cauchy_rnd (@var{location}, @var{scale}, @var{r})
@deftypefnx {Function File} {} cauchy_rnd (@var{location}, @var{scale}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} cauchy_rnd (@var{location}, @var{scale}, [@var{sz}])
Return a matrix of random samples from the Cauchy distribution with
parameters @var{location} and @var{scale}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{location} and @var{scale}.
@end deftypefn


@c chi2rnd scripts/statistics/distributions/chi2rnd.m
@anchor{XREFchi2rnd}
@deftypefn  {Function File} {} chi2rnd (@var{n})
@deftypefnx {Function File} {} chi2rnd (@var{n}, @var{r})
@deftypefnx {Function File} {} chi2rnd (@var{n}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} chi2rnd (@var{n}, [@var{sz}])
Return a matrix of random samples from the chi-square distribution with
@var{n} degrees of freedom.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{n}.
@end deftypefn


@c discrete_rnd scripts/statistics/distributions/discrete_rnd.m
@anchor{XREFdiscrete_rnd}
@deftypefn  {Function File} {} discrete_rnd (@var{v}, @var{p})
@deftypefnx {Function File} {} discrete_rnd (@var{v}, @var{p}, @var{r})
@deftypefnx {Function File} {} discrete_rnd (@var{v}, @var{p}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} discrete_rnd (@var{v}, @var{p}, [@var{sz}])
Return a matrix of random samples from the univariate distribution which
assumes the values in @var{v} with probabilities @var{p}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{v} and @var{p}.
@end deftypefn


@c empirical_rnd scripts/statistics/distributions/empirical_rnd.m
@anchor{XREFempirical_rnd}
@deftypefn  {Function File} {} empirical_rnd (@var{data})
@deftypefnx {Function File} {} empirical_rnd (@var{data}, @var{r})
@deftypefnx {Function File} {} empirical_rnd (@var{data}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} empirical_rnd (@var{data}, [@var{sz}])
Return a matrix of random samples from the empirical distribution obtained
from the univariate sample @var{data}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is a random ordering
of the sample @var{data}.
@end deftypefn


@c exprnd scripts/statistics/distributions/exprnd.m
@anchor{XREFexprnd}
@deftypefn  {Function File} {} exprnd (@var{lambda})
@deftypefnx {Function File} {} exprnd (@var{lambda}, @var{r})
@deftypefnx {Function File} {} exprnd (@var{lambda}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} exprnd (@var{lambda}, [@var{sz}])
Return a matrix of random samples from the exponential distribution with
mean @var{lambda}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{lambda}.
@end deftypefn


@c frnd scripts/statistics/distributions/frnd.m
@anchor{XREFfrnd}
@deftypefn  {Function File} {} frnd (@var{m}, @var{n})
@deftypefnx {Function File} {} frnd (@var{m}, @var{n}, @var{r})
@deftypefnx {Function File} {} frnd (@var{m}, @var{n}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} frnd (@var{m}, @var{n}, [@var{sz}])
Return a matrix of random samples from the F distribution with
@var{m} and @var{n} degrees of freedom.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{m} and @var{n}.
@end deftypefn


@c gamrnd scripts/statistics/distributions/gamrnd.m
@anchor{XREFgamrnd}
@deftypefn  {Function File} {} gamrnd (@var{a}, @var{b})
@deftypefnx {Function File} {} gamrnd (@var{a}, @var{b}, @var{r})
@deftypefnx {Function File} {} gamrnd (@var{a}, @var{b}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} gamrnd (@var{a}, @var{b}, [@var{sz}])
Return a matrix of random samples from the Gamma distribution with
shape parameter @var{a} and scale @var{b}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{a} and @var{b}.
@end deftypefn


@c geornd scripts/statistics/distributions/geornd.m
@anchor{XREFgeornd}
@deftypefn  {Function File} {} geornd (@var{p})
@deftypefnx {Function File} {} geornd (@var{p}, @var{r})
@deftypefnx {Function File} {} geornd (@var{p}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} geornd (@var{p}, [@var{sz}])
Return a matrix of random samples from the geometric distribution with
parameter @var{p}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{p}.

The geometric distribution models the number of failures (@var{x}-1) of a
Bernoulli trial with probability @var{p} before the first success (@var{x}).
@end deftypefn


@c hygernd scripts/statistics/distributions/hygernd.m
@anchor{XREFhygernd}
@deftypefn  {Function File} {} hygernd (@var{t}, @var{m}, @var{n})
@deftypefnx {Function File} {} hygernd (@var{t}, @var{m}, @var{n}, @var{r})
@deftypefnx {Function File} {} hygernd (@var{t}, @var{m}, @var{n}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} hygernd (@var{t}, @var{m}, @var{n}, [@var{sz}])
Return a matrix of random samples from the hypergeometric distribution
with parameters @var{t}, @var{m}, and @var{n}.

The parameters @var{t}, @var{m}, and @var{n} must be positive integers
with @var{m} and @var{n} not greater than @var{t}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{t}, @var{m}, and @var{n}.
@end deftypefn


@c laplace_rnd scripts/statistics/distributions/laplace_rnd.m
@anchor{XREFlaplace_rnd}
@deftypefn  {Function File} {} laplace_rnd (@var{r})
@deftypefnx {Function File} {} laplace_rnd (@var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} laplace_rnd ([@var{sz}])
Return a matrix of random samples from the Laplace distribution.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.
@end deftypefn


@c logistic_rnd scripts/statistics/distributions/logistic_rnd.m
@anchor{XREFlogistic_rnd}
@deftypefn  {Function File} {} logistic_rnd (@var{r})
@deftypefnx {Function File} {} logistic_rnd (@var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} logistic_rnd ([@var{sz}])
Return a matrix of random samples from the logistic distribution.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.
@end deftypefn


@c lognrnd scripts/statistics/distributions/lognrnd.m
@anchor{XREFlognrnd}
@deftypefn  {Function File} {} lognrnd (@var{mu}, @var{sigma})
@deftypefnx {Function File} {} lognrnd (@var{mu}, @var{sigma}, @var{r})
@deftypefnx {Function File} {} lognrnd (@var{mu}, @var{sigma}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} lognrnd (@var{mu}, @var{sigma}, [@var{sz}])
Return a matrix of random samples from the lognormal distribution with
parameters @var{mu} and @var{sigma}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{mu} and @var{sigma}.
@end deftypefn


@c nbinrnd scripts/statistics/distributions/nbinrnd.m
@anchor{XREFnbinrnd}
@deftypefn  {Function File} {} nbinrnd (@var{n}, @var{p})
@deftypefnx {Function File} {} nbinrnd (@var{n}, @var{p}, @var{r})
@deftypefnx {Function File} {} nbinrnd (@var{n}, @var{p}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} nbinrnd (@var{n}, @var{p}, [@var{sz}])
Return a matrix of random samples from the negative binomial distribution
with parameters @var{n} and @var{p}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{n} and @var{p}.
@end deftypefn


@c normrnd scripts/statistics/distributions/normrnd.m
@anchor{XREFnormrnd}
@deftypefn  {Function File} {} normrnd (@var{mu}, @var{sigma})
@deftypefnx {Function File} {} normrnd (@var{mu}, @var{sigma}, @var{r})
@deftypefnx {Function File} {} normrnd (@var{mu}, @var{sigma}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} normrnd (@var{mu}, @var{sigma}, [@var{sz}])
Return a matrix of random samples from the normal distribution with
parameters mean @var{mu} and standard deviation @var{sigma}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{mu} and @var{sigma}.
@end deftypefn


@c poissrnd scripts/statistics/distributions/poissrnd.m
@anchor{XREFpoissrnd}
@deftypefn  {Function File} {} poissrnd (@var{lambda})
@deftypefnx {Function File} {} poissrnd (@var{lambda}, @var{r})
@deftypefnx {Function File} {} poissrnd (@var{lambda}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} poissrnd (@var{lambda}, [@var{sz}])
Return a matrix of random samples from the Poisson distribution with
parameter @var{lambda}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{lambda}.
@end deftypefn


@c stdnormal_rnd scripts/statistics/distributions/stdnormal_rnd.m
@anchor{XREFstdnormal_rnd}
@deftypefn  {Function File} {} stdnormal_rnd (@var{r})
@deftypefnx {Function File} {} stdnormal_rnd (@var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} stdnormal_rnd ([@var{sz}])
Return a matrix of random samples from the standard normal distribution
(mean = 0, standard deviation = 1).

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.
@end deftypefn


@c trnd scripts/statistics/distributions/trnd.m
@anchor{XREFtrnd}
@deftypefn  {Function File} {} trnd (@var{n})
@deftypefnx {Function File} {} trnd (@var{n}, @var{r})
@deftypefnx {Function File} {} trnd (@var{n}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} trnd (@var{n}, [@var{sz}])
Return a matrix of random samples from the t (Student) distribution with
@var{n} degrees of freedom.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{n}.
@end deftypefn


@c unidrnd scripts/statistics/distributions/unidrnd.m
@anchor{XREFunidrnd}
@deftypefn  {Function File} {} unidrnd (@var{n})
@deftypefnx {Function File} {} unidrnd (@var{n}, @var{r})
@deftypefnx {Function File} {} unidrnd (@var{n}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} unidrnd (@var{n}, [@var{sz}])
Return a matrix of random samples from the discrete uniform distribution
which assumes the integer values 1--@var{n} with equal probability.

@var{n} may be a scalar or a multi-dimensional array.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the size of
@var{n}.
@end deftypefn


@c unifrnd scripts/statistics/distributions/unifrnd.m
@anchor{XREFunifrnd}
@deftypefn  {Function File} {} unifrnd (@var{a}, @var{b})
@deftypefnx {Function File} {} unifrnd (@var{a}, @var{b}, @var{r})
@deftypefnx {Function File} {} unifrnd (@var{a}, @var{b}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} unifrnd (@var{a}, @var{b}, [@var{sz}])
Return a matrix of random samples from the uniform distribution on
[@var{a}, @var{b}].

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{a} and @var{b}.
@end deftypefn


@c wblrnd scripts/statistics/distributions/wblrnd.m
@anchor{XREFwblrnd}
@deftypefn  {Function File} {} wblrnd (@var{scale}, @var{shape})
@deftypefnx {Function File} {} wblrnd (@var{scale}, @var{shape}, @var{r})
@deftypefnx {Function File} {} wblrnd (@var{scale}, @var{shape}, @var{r}, @var{c}, @dots{})
@deftypefnx {Function File} {} wblrnd (@var{scale}, @var{shape}, [@var{sz}])
Return a matrix of random samples from the Weibull distribution with
parameters @var{scale} and @var{shape}.

When called with a single size argument, return a square matrix with
the dimension specified.  When called with more than one scalar argument the
first two arguments are taken as the number of rows and columns and any
further arguments specify additional matrix dimensions.  The size may also
be specified with a vector of dimensions @var{sz}.

If no size arguments are given then the result matrix is the common size of
@var{scale} and @var{shape}.
@end deftypefn


@c wienrnd scripts/statistics/distributions/wienrnd.m
@anchor{XREFwienrnd}
@deftypefn {Function File} {} wienrnd (@var{t}, @var{d}, @var{n})
Return a simulated realization of the @var{d}-dimensional Wiener Process
on the interval [0, @var{t}].

If @var{d} is omitted, @var{d} = 1 is used.  The first column of the
return matrix contains time, the remaining columns contain the Wiener
process.

The optional parameter @var{n} defines the number of summands used for
simulating the process over an interval of length 1.  If @var{n} is
omitted, @var{n} = 1000 is used.
@end deftypefn