similar to: power.t.test needs to check delta==NULL before abs(delta) (PR#3139)

It's the same as PR#2993, fixed yesterday in R-patched & R-devel On Fri, 30 May 2003 rossini@blindglobe.net wrote: > > Something like the following should be done for power.t.test to make > sure that it doesn't try to evaluate abs(NULL), which results in an > error. That's a backwards patch, I presume, or you revert the fix. > --- rossini.power.t.test.R

Greetings, I've tried to update the power.t.test function to allow for different sample sizes and sample variances in the case of a two-sample t test. I'd gladly update the corresponding help page if the code is to replace the current power.t.test function. The modified power.t.test code is included below. The changes are as follows: - Added three new arguments to the function *

Dear Sir or Madam, What is the formula for power.t.test(delta=delta, sd=segma, sig.level=0.05, power=0.8, type="one.sample", alternative="one.sided")$n? Thank you very much for the help! Best, Xiaorong

Dear all I am very beginner user of R I want to calculate the power of t-test using power.t.test function. It is stated in the manual that the argument (n) is the number of observations per group. Infact I have a differens number of observations in each group(ie n1<>n2) and another thing is SD the pooled standard deviation of the observations in the two data sets under investigation. Hope

What is the rationale behind having both the delta and sd parameters for the power.t.test function? For the relevant noncentrality parameter, we only need the ratio delta/sd. If my effect size is given as Cohen's d, then I only got that ratio and not sd. As far as I see, in such a case, I can specify delta=d and leave sd at its default value 1. Is this correct or am I missing something?

For various reasons, I spent part of my time today looking at sample size and power calculation tools (don't ask, don't tell...). This seems to be one area that R is incredibly weak in (well, nearly all stat packages, except perhaps specialized tools and SAS); sure, there are a number of functions in various packages: base, statmod, Hmisc Have I missed something? (I

Suppose one wanted to build an R wizard for packaging... One would need to specify and create: package-name/ package-name/Description package-name/INDEX package-name/R/ package-name/man/ package-name/src/ >From this, we'd "probably" like to: - script-up the Description - provide a script for regenerating INDEX by calling out to the

Duncan - (but sent to R-devel, for any other thoughts?) How do you envision RSPerl being used? Without having seen the details, I can think of playing with strings; is there a preferred incantation for stringification of R objects (serialization)? (actually, the main problem I'm having is that I know how I want to code something like: Robject <-

could we add an argument to suppress printing of levels for print.factor()? i.e.: print.factor <- function (x, quote = FALSE, print.levels = TRUE, ...) ## MODIFIED { if (length(x) <= 0) cat("factor(0)\n") else print(as.character(x), quote = quote, ...) if (print.levels) { ## ADDED cat("Levels:

There's a bit of a symmetry issue, which may or may not be important (led to 15 seconds of confusion until I got my bearings straight): row.names, rownames; colnames, no col.names? I _NOW_ realize the difference between row.names and rownames, but is there any reason not to have col.names for re-naming columns in a data.frame? (if there is, I don't particular need to know it, but it

seek(zz,where=NewPosition) returns the OldPosition of connection zz (of the appropriate "pointer", i.e. read or write). Currently, the docs have: > Details: > > `seek' with `where = NA' returns the current byte offset of a > connection (from the beginning), and with a positive `where' > argument the connection is re-positioned (if possible) to

What's the current best approach to dealing with the nlminb problem? Are there other similar constrained optimizers, or is current wisdom just to code it in C (or similar...)? best, -tony -- A.J. Rossini Research Assistant Professor of Biostatistics Biostatistics/Univ. of Washington (Th) Box 357232 206-543-1044 (3286=fax) Center for AIDS Research/HMC/UW (M/F) Box 359931

Thanks to Tony Plate for letting me know what the abbreviate.arg option does. I think this could be made more flexible (I.e. =TRUE, =FALSE, =#, where # would be passed to the abbreviate min.length argument). But it follows the example I was given. "print.factor" <- function (x, quote = FALSE, max.levels=5, print.levels = {if (max.levels==0) FALSE else TRUE},

One thing that has driven me batty to no end is the fact that ESS doesn't handle R's "exit" well. Any thoughts as to the default? I'm thinking of something like: current: do what currently is done C-u Y C-c C-q : quit and save workspace C-u N C-c C-q : quit and don't save workspace. Comments? best, -tony -- A.J. Rossini Rsrch. Asst.

After installing nlme_3.1-7.tar.gz under the latest R (1.2.0, via anonymous CVS), I'm unable to get it to work, i.e.: rossini 94 > /usr/local/bin/R R : Copyright 2000, The R Development Core Team Version 1.2.0 Under development (unstable) (2000-12-06) R is free software and comes with ABSOLUTELY NO WARRANTY. You are welcome to redistribute it under certain conditions. Type

> From: rossini@blindglobe.net (A.J. Rossini) > Date: 01 Nov 2000 07:47:21 -0800 [...] > Thanks for the pointer to stack/unstack -- now, having been reminded, > I think I'd seen these float through on the list (still doesn't solve > the missing modeling routines (parametric GLMMs, some of the > econometrics stuff -- does R _easily_ do 3SLS?), but they'll appear >

Hello, I checked mail list archives and this topic came up before in March '99 (about ability of R to multithread). At that time rossini at biostat.washington.edu (Tom Rossini) suggested to check his web page in a few month as it was on his list to-do. I checked but the web page is not valid. I wanted to know if in general progress has been made in the area, or if it is planned. I was also

Hi to all who suffer from rterm not shutting down in xemacs/ESS on windows NT or 2000. Also hi to those who could eventually help. Here is some more information which could help and some ENCOURAGEMENT to contribute to a solution. 1. It may be an xemacs problem but it is more likely an interaction between rterm/comint/and xemacs. In fact, the problem started occurring around version R 1.6.0. I

Is it possible to run R soley from the shell? I'm evaluating necessary fixes to R and XLispStat (and eventually SAS and S-PLUS 4) for running as subprocesses to Emacs (NTEmacs) for ESS. Currently, I lose control of R when running via Emacs; is it possible to "remove" the windows? (all, or all but the graphics window)? best, -- -tony (Anthony Rossini) Statistics

for Spam. In the process of setting up a more effective spam filtering system, I just noticed that bogofilter, which implements extensions of the (a?) "Naive Bayes" text classification approach, will dump out R data frames; the man page suggests how to "integrate" it with R for verification. (sort of, that is). Anyway, for those of you looking for silly and perhaps