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
|
\name{msfit}
\alias{msfit}
\alias{plot.msfit}
\alias{lines.msfit}
\alias{summary.msfit}
\title{Curve Fitting for Multiscale Bootstrap Resampling}
\description{\code{msfit} performs curve fitting for multiscale
bootstrap resampling. It generates an object of class
\code{msfit}. Several generic methods are available.
}
\usage{
msfit(bp, r, nboot)
\method{plot}{msfit}(x, curve=TRUE, main=NULL, sub=NULL, xlab=NULL, ylab=NULL, ...)
\method{lines}{msfit}(x, col=2, lty=1, ...)
\method{summary}{msfit}(object, digits=3, ...)
}
\arguments{
\item{bp}{numeric vector of bootstrap probability values.}
\item{r}{numeric vector of relative sample size of bootstrap samples
defined as \eqn{r=n'/n} for original sample size \eqn{n} and
bootstrap sample size \eqn{n'}.}
\item{nboot}{numeric value (vector) of the number of bootstrap
replications.}
\item{x}{object of class \code{msfit}.}
\item{curve}{logical. If \code{TRUE}, the fitted curve is drawn.}
\item{main, sub, xlab, ylab, col, lty}{generic graphic parameters.}
\item{object}{object of class \code{msfit}.}
\item{digits}{integer indicating the precision to be used in rounding.}
\item{...}{other parameters to be used in the functions.}
}
\details{
function \code{msfit} performs the curve fitting for multiscale
bootstrap resampling. In package \code{pvclust} this function is only
called from the function \code{pvclust} (or \code{parPvclust}), and
may never be called from users. However one can access a list of
\code{msfit} objects by \code{x$msfit}, where \code{x} is an object of
class \code{pvclust}.
}
\value{\code{msfit} returns an object of class \code{msfit}. It contains
the following objects:
\item{p}{numeric vector of \eqn{p}-values. \code{au} is AU
(Approximately Unbiased) \eqn{p}-value computed by multiscale
bootstrap resampling, which is more accurate than BP value
(explained below) as unbiased \eqn{p}-value. \code{bp} is BP
(Bootstrap Probability) value, which is simple but tends to be
unbiased when the absolute value of \code{c} (a value in \code{coef}
vector, explained below) is large.}
\item{se}{numeric vector of estimated standard errors of \eqn{p}-values.}
\item{coef}{numeric vector related to geometric aspects of
hypotheses. \code{v} is signed distance and \code{c} is curvature of
the boundary.}
\item{df}{numeric value of the degree of freedom in curve fitting.}
\item{rss}{residual sum of squares.}
\item{pchi}{\eqn{p}-value of chi-square test based on asymptotic theory.}
}
\references{
Shimodaira, H. (2004)
"Approximately unbiased tests of regions using multistep-multiscale
bootstrap resampling",
\emph{Annals of Statistics}, 32, 2616-2641.
Shimodaira, H. (2002)
"An approximately unbiased test of phylogenetic tree selection",
\emph{Systematic Biology}, 51, 492-508.
}
\author{Ryota Suzuki \email{suzuki@ef-prime.com}}
\keyword{htest}
|
