[R] use of predict() with confidence/prediction bands
To my understanding, a confidence interval typically covers a single valued parameter. In contrast, a confidence band covers an entire line with a band. In regression, it is quite common to construct confidence and prediction bands. I have found that many people are connecting individual confidence/prediction interval values produced with predict(object,sd.fit=T,type=conf/pred) and calling the result a confidence/prediction band. Since there is no specific probability statement that can be attached to these connected confidence/prediction intervals, this does not seem reasonable to me. This is done, for example, in ISWR pg. 105, UsingR for Introductory Statistics pg 296, and Linear Models with R pg. 39 (Although in this instance the intervals are called 95% pointwise confidence bands versus simply 95% confidence bands.) To make a confidence/prediction band, one should construct simultaneous confidence/prediction intervals with say a Scheffe approach as mentioned in the S-PLUS Guide to statistics pg 274. If these connected intervals were called pointwise confidence/prediction intervals with the understanding that have no particular probability interpretation, then they are useful in understanding where the line should fall. However, they are not confidence/prediction bands as such, and I think it is misleading to name them so. Should the intervals the authors in the three mentioned references construct not be called something similar to connected 95% pointwise confidence/prediction intervals versus 95% confidence/prediction bands? Or, have I missed the boat? Fire away... Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Creating packages with windows (accessing data)
Dear R List, I have created a package (under Windows 2.0.1) with 300+ data sets and 20 or so functions I use in teaching. However, to access the data, one needs to type data(foo) once the package has been installed and loaded. With other packages namely MASS, after the package is installed and loaded with library(MASS), it is possible to refer to a data set say Animals by simply typing Animals at the command prompt. I would like to have similar functionality in my package. Would someone provide some hints as to what I need to do (read about xxx...or provide a line of code) so that the data in the package can be accessed once it is loaded without typing data(foo) all the time. Currently, when I type the name of a function it shows the code for the function. I would like it to also show the data when I type the name of any of the data sets. Thanks for the pointers in advance. Alan- Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Error Building From Source
Greetings: I am trying to build R-2.0.1 from source on windows. My path is set to: .;C:\RStools;C:\MinGW\bin;C:\perl\bin;C:\texmf\miktex\bin;C:\HTMLws\;C:\R201\R201\bin;%System Root%\system32;%SystemRoot%;%SystemRoot%\System32\Wbem;C:\Program Files\ Common Files\Adaptec Shared\System;C:\LINGO9\ and Mkrules has been edited and reads # path (possibly full path) to same version of R on the host system R_EXE=C:/R201/R201/bin when I type make I get the following: -- Making package base adding build stamp to DESCRIPTION C:/R201/R201/bin: not found make[4]: *** [frontmatter] Error 127 make[3]: *** [all] Error 2 make[2]: *** [pkg-base] Error 2 make[1]: *** [rpackage] Error 2 make: *** [all] Error 2 Any hints as to why it says it can not find C:/R201/R201/bin? Any help would be appreciated. Alan Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Building Windows Package
I am using R-1.9.1 with windows 2000 and trying to build a package. However, when I issue the command: RCMD build --binary BSDA I get: Building/Updating help pages for package 'BSDA' Formats: chm hhc: not found cp: cannot stat `C:/R191/R191/JUNK/BSDA/chm/BSDA.chm': No such file or direc tory make[1]: *** [chm-BSDA] Error 1 make: *** [pkg-BSDA] Error 2 *** Installation of BSDA failed *** My path is: .;C:\RStools;C:\MinGW\bin;C:\perl\bin;C:\texmf\miktex\bin;C:\Rstools\zip.exe; C:\Rstools\unzip.exe;C:\HTMLws\hhc.exe;C:\R191\R191\bin;%SystemRoot%\system32; %SystemRoot%;%SystemRoot%\System32\Wbem;C:\Program Files\CommonFiles\AdaptecShared\System; C:\Program Files\ggobi;C:\Program Files\R\rw1091\library\Rggobi\libs (HTML Help Workshop lives in- C:\HTMLws\hhc.exe) I have the directory for HTML Help workshop in MkRules set as: # Where does 'HTML Help Workshop' live? (unused if compiled HTML help is # not requested. Spaces allowed.) HHWDIR=C:/HTMLws After I issue RCMD build --binary BSDA and get the error messages, if I go to `C:/R191/R191/JUNK/BSDA/chm/BSDA.hhp' and manually compile the file (with HTML Help workshop) then issue the command RCMD build --binary BSDA again...the package builds without problems. Any ideas what I am doing wrong? Thanks in advance for the help. Alan Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University 2003-2004 International Exchange Scholar Universidad Publica de Navarra - Nafarroako Unibertsitate Publikoa Pamplona, Spain TEL : +34 948 169 205 CELL: +34 656 668 621 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Building Windows Package
Thanks. It works great now! Alan Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University 2003-2004 International Exchange Scholar Universidad Publica de Navarra - Nafarroako Unibertsitate Publikoa Pamplona, Spain TEL : +34 948 169 205 CELL: +34 656 668 621 On Mon, 26 Jul 2004, Prof Brian Ripley wrote: You want the directory in your path specification, not the executables. As in .;C:\RStools;C:\MinGW\bin;C:\perl\bin;C:\texmf\miktex\bin; C:\HTMLws;C:\R191\R191\bin;%SystemRoot%\system32; %SystemRoot%;%SystemRoot%\System32\Wbem;C:\Program Files\CommonFiles\AdaptecShared\System; C:\Program Files\ggobi;C:\Program Files\R\rw1091\library\Rggobi\libs On Mon, 26 Jul 2004, Alan Arnholt wrote: I am using R-1.9.1 with windows 2000 and trying to build a package. However, when I issue the command: RCMD build --binary BSDA I get: Building/Updating help pages for package 'BSDA' Formats: chm hhc: not found cp: cannot stat `C:/R191/R191/JUNK/BSDA/chm/BSDA.chm': No such file or direc tory make[1]: *** [chm-BSDA] Error 1 make: *** [pkg-BSDA] Error 2 *** Installation of BSDA failed *** My path is: .;C:\RStools;C:\MinGW\bin;C:\perl\bin;C:\texmf\miktex\bin;C:\Rstools\zip.exe; C:\Rstools\unzip.exe;C:\HTMLws\hhc.exe;C:\R191\R191\bin;%SystemRoot%\system32; %SystemRoot%;%SystemRoot%\System32\Wbem;C:\Program Files\CommonFiles\AdaptecShared\System; C:\Program Files\ggobi;C:\Program Files\R\rw1091\library\Rggobi\libs (HTML Help Workshop lives in- C:\HTMLws\hhc.exe) I have the directory for HTML Help workshop in MkRules set as: # Where does 'HTML Help Workshop' live? (unused if compiled HTML help is # not requested. Spaces allowed.) HHWDIR=C:/HTMLws After I issue RCMD build --binary BSDA and get the error messages, if I go to `C:/R191/R191/JUNK/BSDA/chm/BSDA.hhp' and manually compile the file (with HTML Help workshop) then issue the command RCMD build --binary BSDA again...the package builds without problems. Any ideas what I am doing wrong? Thanks in advance for the help. Alan Alan T. Arnholt Associate Professor Dept. of Mathematical Sciences Appalachian State University 2003-2004 International Exchange Scholar Universidad Publica de Navarra - Nafarroako Unibertsitate Publikoa Pamplona, Spain TEL : +34 948 169 205 CELL: +34 656 668 621 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Exact Binomial test feature or bug?
Dear R Users,
Is the p-value reported in a two-tailed binomial exact
test in error or is it a feature?
If it is a feature, could someone provide a reference
for its two-tailed p-value computations?
Using Blaker's (2000 - Canad. J. Statist 28: 783-798)
approach,the p-value is the minimum of the two-tailed
probabilities $P \left(Y\geq y_{obs}\right)$ and
$P\left(Y\leq y_{obs}\right)$ plus an attainable
probability in the other tail that is as close as
possible to, but not greater than that one-tailed
probability. Consider the following examples
performed in R version 1.9
under windows 2000.
binom.test(110,500,.2)
Exact binomial test
data: 110 and 500
number of successes = 110, number of trials = 500, p-
value = 0.2636
alternative hypothesis: true probability of success is
not equal to 0.2
95 percent confidence interval:
0.1844367 0.2589117
sample estimates:
probability of success
0.22
(P-value should be according to the Blaker definition
0.2881)
binom.test(90,500,.2)
Exact binomial test
data: 90 and 500
number of successes = 90, number of trials = 500, p-
value = 0.2881
alternative hypothesis: true probability of success is
not equal to 0.2
95 percent confidence interval:
0.1473006 0.2165364
sample estimates:
probability of success
0.18
(P-value should be according to the Blaker definition
0.2647)
The following code (only checked on windows) reports
the Blaker definition of the p-value.
%%%
bino.test -
function(x, n, p = 0.5, alternative = c
(two.sided, less, greater), conf.level = 0.95)
{
eps - sqrt(.Machine$double.eps)
if(any(is.na(x) || (x 0) || (x != round(x
stop(x must be nonnegative and
integer)
if(length(x) == 2) {
n - sum(x)
x - x[1]
}
else if(length(x) == 1) {
if((length(n) 1) || is.na(n) || (n
1) || (n != round(n)) || (x n))
stop(n must be a positive
integer = x)
}
else stop(incorrect length of x)
if(!missing(p) (length(p) 1 || is.na(p)
|| p 0 || p 1))
stop(p must be a single number
between 0 and 1)
alternative - match.arg(alternative)
if(!((length(conf.level) == 1) is.finite
(conf.level) (conf.level 0)
(conf.level 1)))
stop(conf.level must be a single
number between 0 and 1)
DNAME - paste(deparse(substitute(x)), and,
deparse(substitute(n)))
PVAL - switch(alternative,
less = pbinom(x, n, p),
greater = 1 - pbinom(x - 1, n, p),
{
pvec - pbinom(0:n, n, p)
if(x/n p) {
pb - pbinom(x, n, p)
p2 - max((1 - pvec)
[pb - (1 - pvec) = - eps * pb], 0)
min(1, pb + p2)
}
else {
pb - (1 - pbinom(x -
1, n, p))
p2 - max(pvec[pb -
pvec = - eps * pb], 0)
min(1, pb + p2)
}
}
)
p.L - function(x, n, alpha)
{
if(x == 0)
0
else qbeta(alpha, x, n - x + 1)
}
p.U - function(x, n, alpha)
{
if(x == n)
1
else qbeta(1 - alpha, x + 1, n - x)
}
CINT - switch(alternative,
less = c(0, p.U(x, n, 1 - conf.level)),
greater = c(p.L(x, n, 1 - conf.level),
1),
two.sided = {
alpha - (1 - conf.level)/2
c(p.L(x, n, alpha), p.U(x, n,
alpha))
}
)
attr(CINT, conf.level) - conf.level
ESTIMATE - x/n
names(x) - number of successes
names(n) - number of trials
names(ESTIMATE) - names(p) - probability of
success
structure(list(statistic = x, parameter = n,
p.value = PVAL, conf.int = CINT, estimate =
ESTIMATE,
null.value = p, alternative =
alternative, method = Exact binomial test,
data.name = DNAME),
class = htest)
}
%%%
%%
Thanks in advance for the responses. I am not
currently on the R-mailing list. So, if you could
cc your responses to [EMAIL PROTECTED] I would
be most thankful.
Alan Arnholt
-
This message was sent using Endymion MailMan.
http://www.endymion.com/products/mailman/
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
