https://github.com/cran/Hmisc
Tip revision: 462397b7fd57a4cb3bf0b37c0fcceb0b208299b8 authored by Charles Dupont on 02 May 2007, 00:00:00 UTC
version 3.4-2
version 3.4-2
Tip revision: 462397b
Misc.Rd
\name{Misc}
\alias{clowess}
\alias{confbar}
\alias{getLatestSource}
\alias{inverseFunction}
\alias{james.stein}
\alias{km.quick}
\alias{lm.fit.qr.bare}
\alias{matxv}
\alias{nomiss}
\alias{outerText}
\alias{sepUnitsTrans}
\alias{trap.rule}
\alias{trellis.strip.blank}
\alias{under.unix}
\alias{.R.}
\alias{.SV4.}
\alias{unPaste}
\alias{whichClosest}
\alias{whichClosePW}
\alias{xless}
\title{Miscellaneous Functions}
\description{
This documents miscellaneous small functions in Hmisc that may be of
interest to users.
\code{clowess} runs \code{lowess} but if the \code{iter} argument
exceeds zero, sometimes wild values can result, in which case
\code{lowess} is re-run with \code{iter=0}.
\code{confbar} draws multi-level confidence bars using small rectangles
that may be of different colors.
\code{getLatestSource} fetches and \code{source}s the most recent
source code for functions in packages in the Vanderbilty University
CVS repository.
\code{inverseFunction} generates a function to find all inverses of a
monotonic or nonmonotonic function that is tabulated at vectors (x,y),
typically 1000 points. If the original function is monotonic, simple linear
interpolation is used and the result is a vector, otherwise linear
interpolation is used within each interval in which the function is
monotonic and the result is a matrix with number of columns equal to the
number of monotonic intervals. If a requested y is not within any
interval, the extreme x that pertains to the nearest extreme y is
returned. Specifying what='sample' to the returned function will cause a
vector to be returned instead of a matrix, with elements taken as a
random choice of the possible inverses.
\code{james.stein} computes James-Stein shrunken estimates of cell
means given a response variable (which may be binary) and a grouping
indicator.
\code{km.quick} provides a fast way to invoke \code{survfit.km} in the
\code{survival} package to get Kaplan-Meier estimates for a
single stratum for a vector of time points (if \code{times} is given) or to
get a vector of survival time quantiles (if \code{q} is given).
\code{lm.fit.qr.bare} is a fast stripped-down function for computing
regression coefficients, residuals, \eqn{R^2}, and fitted values. It
uses the Fortran routines \code{dqrls}.
\code{matxv} multiplies a matrix by a vector, handling automatic
addition of intercepts if the matrix does not have a column of ones.
If the first argument is not a matrix, it will be converted to one.
\code{nomiss} returns a data frame (if its argument is one) with rows
corresponding to \code{NA}s removed, or it returns a matrix with rows
with any element missing removed.
\code{outerText} uses \code{text()} to put test strings in left or
right margins. It temporarily sets \code{par(xpd=NA)} if using \R.
\code{sepUnitsTrans} converts character vectors containing values such
as \code{c("3 days","3day","4month","2 years","2weeks","7")} to
numeric vectors
(here \code{c(3,3,122,730,14,7)}) in a flexible fashion. The user can
specify a
vector of units of measurements and conversion factors. The units
with a conversion factor of \code{1} are taken as the target units,
and if those units are present in the character strings they are
ignored. The target units are added to the resulting vector as the
\code{"units"} attribute.
\code{trap.rule} computes the area under a curve using the trapezoidal
rule, assuming \code{x} is sorted.
\code{trellis.strip.blank} sets up Trellis or Lattice graphs to have a
clear background on the strips for panel labels.
\code{under.unix} is a scalar logical value that is \code{TRUE} if you
are running Linux or Unix.
\code{.R.} is a logical value set to \code{TRUE} if running \R,
\code{FALSE} for S-Plus.
\code{.SV4.} is a logical value set to \code{TRUE} if running version
4 of the S language under S-Plus, \code{FALSE} otherwise.
\code{unPaste} provides a version of the S-Plus \code{unpaste} that
works for \R and S-Plus.
\code{whichClosePW} is a very fast function using weighted multinomial
sampling to determine which element of a vector is "closest" to each
element of another vector. \code{whichClosest} quickly finds the closest
element without any randomness.
\code{xless} is a function for Linux/Unix users to invoke the system
\code{xless} command to pop up a window to display the result of
\code{print}ing an object. For Windows, \code{xless} uses the builtin
\code{page} function but with better defaults.
}
\usage{
confbar(at, est, se, width, q = c(0.7, 0.8, 0.9, 0.95, 0.99),
col = if (.R.) gray(c(0, 0.25, 0.5, 0.75, 1)) else
if (under.unix) c(1, 0.8, 0.5, 0.2, 0.065) else
c(1, 4, 3, 2, 5),
type = c("v", "h"), labels = TRUE, ticks = FALSE,
cex = 0.5, side = "l", lwd = 5, clip = c(-1e+30, 1e+30),
fun = function(x) x,
qfun = function(x) ifelse(x == 0.5, qnorm(x),
ifelse(x < 0.5, qnorm(x/2),
qnorm((1 + x)/2))))
getLatestSource(x=NULL, package='Hmisc', recent=NULL, avail=FALSE,
type=c('svn','cvs'))
inverseFunction(x, y)
james.stein(y, group)
km.quick(S, times, q)
lm.fit.qr.bare(x, y, tolerance, intercept=TRUE, xpxi=FALSE)
matxv(a, b, kint=1)
nomiss(x)
outerText(string, y, setAside=string[1], side=4, space=1,
adj=1, cex=par('cex'))
sepUnitsTrans(x, conversion=c(day=1, month=365.25/12, year=365.25, week=7),
round=FALSE, digits=0)
trap.rule(x, y)
trellis.strip.blank()
unPaste(str, sep="/", extended=FALSE)
whichClosest(x, w)
whichClosePW(x, w, f=0.2)
xless(x, \dots, title)
}
\arguments{
\item{a}{a numeric matrix or vector}
\item{adj}{0 for left justification, 0.5 for centered, 1 for right}
\item{at}{x-coordinate for vertical confidence intervals, y-coordinate
for horizontal}
\item{avail}{set to \code{TRUE} to have \code{getLatestSource} return
a data frame of available files and latest versions instead of
fetching any}
\item{b}{a numeric vector}
\item{cex}{character expansion factor}
\item{clip}{interval to truncate limits}
\item{col}{vector of colors}
\item{conversion}{a named numeric vector}
\item{digits}{number of digits used for \code{round}}
\item{est}{vector of point estimates for confidence limits}
\item{extended}{see \code{\link{strsplit}} in \R}
\item{f}{a scaling constant}
\item{fun}{function to transform scale}
\item{group}{a categorical grouping variable}
\item{intercept}{set to \code{FALSE} to not automatically add a column
of ones to the \code{x} matrix}
\item{kint}{which element of \code{b} to add to the result if \code{a}
does not contain a column for intercepts}
\item{labels}{set to \code{FALSE} to omit drawing confidence
coefficients}
\item{lwd}{line widths}
\item{package}{name of package for \code{getLatestSource}, default is
\code{'Hmisc'}}
\item{q}{vector of confidence coefficients or quantiles}
\item{qfun}{quantiles on transformed scale}
\item{recent}{an integer telling \code{getLatestSource} to get the
\code{recent} most recently modified files from the package}
\item{round}{set to \code{TRUE} to round converted values}
\item{S}{a \code{\link[survival]{Surv}} object}
\item{se}{vector of standard errors}
\item{sep}{a single character string specifying the delimiter}
\item{setAside}{for \code{adj=1 side=4}, is a character string used to
determine the space to set aside for all strings.}
\item{side}{for \code{confbar} is \code{"b","l","t","r"} for bottom,
left, top, right. For \code{outText} is the using integers 1-4
corresponding to these.}
\item{space}{the number of extra characters to leave to the left of
the string(s) (\code{adj=0}) or to the right (\code{adj=1})}
\item{str}{a character string vector}
\item{string}{a character string vector}
\item{ticks}{set to \code{TRUE} to draw lines between rectangles}
\item{times}{a numeric vector of times}
\item{title}{a character string to title a window or plot}
\item{tolerance}{tolerance for judging singularity in matrix}
\item{type}{\code{"v"} for vertical, \code{"h"} for horizontal. For
\code{getLatestSource} this specifies the type of source code
repository, \code{'svn'} (the default) or
\code{'cvs'}, which is now outdated as Subversion has replaced CVS
in the Vanderbilt Biostatistics server.}
\item{w}{a numeric vector}
\item{width}{width of confidence rectanges in user units}
\item{x}{a numeric vector (matrix for \code{lm.fit.qr.bare}) or data
frame. For \code{xless} may be any object that is sensible to
\code{print}. For \code{sepUnitsTrans} is a character or factor
variable. For \code{getLatestSource} is a character string or
vector of character strings containing base file names to retrieve
from CVS. Set \code{x='all'} to retrieve all source files. For
\code{clowess}, \code{x} may also be a list with x and y
components. For \code{inverseFunction}, \code{x} and \code{y}
contain evaluations of the function whose inverse is needed.
\code{x} is typically an equally-spaced grid of 1000 points.}
\item{xpxi}{set to \code{TRUE} to add an element to the result
containing the inverse of \eqn{X'X}}
\item{y}{a numeric vector. For \code{inverseFunction} \code{y} is the
evaluated function values at \code{x}.}
\item{\dots}{arguments passed through to another function}
}
\author{Frank Harrell}
\examples{
trap.rule(1:100,1:100)
unPaste(c('a;b or c','ab;d','qr;s'), ';')
sepUnitsTrans(c('3 days','4 months','2 years','7'))
set.seed(1)
whichClosest(1:100, 3:5)
whichClosest(1:100, rep(3,20))
whichClosePW(1:100, rep(3,20))
whichClosePW(1:100, rep(3,20), f=.05)
whichClosePW(1:100, rep(3,20), f=1e-10)
x <- seq(-1, 1, by=.01)
y <- x^2
h <- inverseFunction(x,y)
formals(h)$turns # vertex
a <- seq(0, 1, by=.01)
plot(0, 0, type='n', xlim=c(-.5,1.5))
lines(a, h(a)[,1]) ## first inverse
lines(a, h(a)[,2], col='red') ## second inverse
a <- c(-.1, 1.01, 1.1, 1.2)
points(a, h(a)[,1])
\dontrun{
getLatestSource(recent=5) # source() most recent 5 revised files in Hmisc
getLatestSource('cut2') # fetch and source latest cut2.s
getLatestSource('all') # get everything
getLatestSource(avail=TRUE) # list available files and latest versions
}
}
\keyword{programming}
\keyword{utilities}