1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120

Probability and Statistics
Probability
*INTRO
Each distribution is represented as an entity. For each distribution
known to the system the consistency of parameters is checked. If
the parameters for a distribution are invalid, the functions return
{Undefined}.
For example, {NormalDistribution}{(a,1)} evaluates to
{Undefined}, because of negative variance.
*CMD BernoulliDistribution  Bernoulli distribution
*STD
*CALL
BernoulliDistribution(p)
*PARMS
{p}  number, probability of an event in a single trial
*DESC
A random variable has a Bernoulli distribution with probability {p} if
it can be interpreted as an indicator of an event, where {p} is the
probability to observe the event in a single trial.
Numerical value of {p} must satisfy $0<p<1$.
*SEE BinomialDistribution
*CMD BinomialDistribution  binomial distribution
*STD
*CALL
BinomialDistribution(p,n)
*PARMS
{p}  number, probability to observe an event in single trial
{n}  number of trials
*DESC
Suppose we repeat a trial {n} times, the probability to observe an
event in a single trial is {p} and outcomes in all trials are mutually
independent. Then the number of trials when the event occurred
is distributed according to the binomial distribution. The probability
of that is {BinomialDistribution}{(p,n)}.
Numerical value of {p} must satisfy $0<p<1$. Numerical value
of {n} must be a positive integer.
*SEE BernoulliDistribution
*CMD tDistribution  Student's $t$ distribution
*STD
*CALL
{tDistribution}(m)
*PARMS
{m}  integer, number of degrees of freedom
*DESC
*REM what does it do???
The function {tDistribution} returns the ...
Let $Y$ and $Z$ be independent random variables, $Y$ have the
NormalDistribution(0,1), {Z} have ChiSquareDistribution(m). Then
$Y/Sqrt(Z/m)$ has tDistribution(m).
Numerical value of {m} must be positive integer.
*CMD PDF  probability density function
*STD
*CALL
PDF(dist,x)
*PARMS
{dist}  a distribution type
{x}  a value of random variable
*DESC
If {dist} is a discrete distribution, then {PDF} returns the
probability for a random variable with distribution {dist} to take a
value of {x}. If {dist} is a continuous distribution, then {PDF}
returns the density function at point $x$.
*SEE CDF
Statistics
*CMD ChiSquareTest  Pearson's ChiSquare test
*STD
*CALL
ChiSquareTest(observed,expected)
ChiSquareTest(observed,expected,params)
*PARMS
{observed}  list of observed frequencies
{expected}  list of expected frequencies
{params}  number of estimated parameters
*DESC
{ChiSquareTest} is intended to find out if our sample was drawn from a
given distribution or not. To find this out, one has to calculate
observed frequencies into certain intervals and expected ones. To
calculate expected frequency the formula $n[i]:=n*p[i]$ must be used,
where $p[i]$ is the probability measure of $i$th interval, and $n$ is
the total number of observations. If any of the parameters of the
distribution were estimated, this number is given as
{params}.
The function returns a list of three local substitution rules. First
of them contains the test statistic, the second contains the value of the parameters, and
the last one contains the degrees of freedom.
The test statistic is distributed as ChiSquareDistribution.
