Mail Archive Search: <+from:naomi@stat.psu.edu>

Results:

References: [ +from:naomi@stat.psu.edu: 21 ]

Total 21 documents matching your query.

1. nested lme model (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Sun, 12 Oct 2003 00:42:24 -0400; I cannot figure out how to set up an unbalanced block design as follows: Arrays and samples are random. There are 2 samples on each array, but each sample may be on 2 or more arrays. Spots within arr; /archives/html/s-news/2003-10/msg00121.html (7,090 bytes)
2. Re: Help with while() (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Thu, 10 Jul 2003 14:54:22 -0400; If you want to print in a loop (rather than save all the values) use print. i.e. sum.so.far <- n <- 0 + n <- n + 1 + sum.so.far <- c(sum.so.far, sum.so.far[k] + n) print(sum.so.far) Alain Yamakana wr; /archives/html/s-news/2003-07/msg00072.html (9,371 bytes)
3. Re: model performance with and without interactions... (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Fri, 20 Jun 2003 11:55:36 -0400; I would not drop interactions based on the relative size of F-tests. I look at the interaction plots, as suggested here. Some statisticians would NEVER drop interactions, because there are inferentia; /archives/html/s-news/2003-06/msg00148.html (13,178 bytes)
4. Re: Question on generalized linear models (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Thu, 03 Apr 2003 13:58:19 -0500; m; /archives/html/s-news/2003-04/msg00022.html (10,272 bytes)
5. Re: lme question (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Thu, 17 Apr 2003 14:29:40 -0400; n; /archives/html/s-news/2003-04/msg00119.html (8,583 bytes)
6. Re: vectors definition and storage (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 22 Apr 2003 10:50:57 -0400; e; /archives/html/s-news/2003-04/msg00136.html (9,934 bytes)
7. Re: Linear model question: independence? (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Thu, 27 Mar 2003 17:26:05 -0500; ies from; /archives/html/s-news/2003-03/msg00198.html (9,398 bytes)
8. Re: microarray power and biological variability (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Fri, 01 Nov 2002 16:52:45 -0500; While we are on this topic, it should be noted that the 2 microarrays per subject should probably be done with a dye-swap. This is because dye-bias can be an important effect on the arrays. Two good; /archives/html/s-news/2002-11/msg00005.html (12,996 bytes)
9. Linking Fortran code (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Mon, 11 Nov 2002 09:21:13 -0500; One of our graduate students asked me to forward this message: Hi all, I need to link a f77 code with splus. Under Unix, Splus3.4 allows for dyn.load function. But, I am using an old code and one of; /archives/html/s-news/2002-11/msg00093.html (7,587 bytes)
10. Re: heterogenous variance of random effect in lme (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Wed, 13 Nov 2002 09:05:45 -0500; The random statement allows you to enter a model. Then the variance depends on the level of the model. Alternatively, you can used pdBlocked to set up a block diagonal random effects variance matrix.; /archives/html/s-news/2002-11/msg00121.html (9,634 bytes)
11. documentation (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 01 Oct 2002 14:17:51 -0400; I have an Splus Reference Manual dating back to 1991 which contains a set of indices to the commands: alphabetical permuted index by function short description by type - e.g. matrix functions Is this; /archives/html/s-news/2002-10/msg00007.html (6,748 bytes)
12. documentation (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 01 Oct 2002 14:48:11 -0400; Browsing around on the Unix system, I found the help.start() command in Splus, which gives me exactly what I want in terms of indices to the documentation. Naomi S. Altman 814-865-3791 (voice) Associ; /archives/html/s-news/2002-10/msg00008.html (6,589 bytes)
13. variance components (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Fri, 11 Oct 2002 09:21:06 -0400; I am fitting an nonlinear mixed model. The model has 4 random effects. Let's call the output from this model "nlme.out". I fit using ML. 1) summary(nlme.out) prints the 4 estimated variance component; /archives/html/s-news/2002-10/msg00094.html (7,172 bytes)
14. Variance components question 1 (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Fri, 11 Oct 2002 10:46:07 -0400; Thanks to Sundar Dorai-Raj of PDF Solutions who provided the following code to extract the estimated variance components from an nlme object: extract.stdev <- function(object) { struct <- object$mode; /archives/html/s-news/2002-10/msg00096.html (6,911 bytes)
15. lme "start" (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 29 Oct 2002 10:51:48 -0500; Is it possible to enter initial values for the random effects in lme? (I know this can be done in nlme, and I will go this route if I must.) Naomi S. Altman 814-865-3791 (voice) Associate Professor D; /archives/html/s-news/2002-10/msg00248.html (6,450 bytes)
16. lme random effects (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 29 Oct 2002 11:26:07 -0500; Is it possible to enter initial values for the random effects in lme? (I know this can be done in nlme, and I will go this route if I must.) Naomi S. Altman 814-865-3791 (voice) Associate Professor D; /archives/html/s-news/2002-10/msg00249.html (6,464 bytes)
17. lme problem (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Thu, 05 Sep 2002 12:33:35 -0400; We are doing a simulation that calls lme. In each call we create a new data set. Usually the code works well but occasionally we get this error message: Problem in .C("mixed_loglik", subroutine mixed; /archives/html/s-news/2002-09/msg00022.html (6,785 bytes)
18. nlme + regression object (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Mon, 05 Aug 2002 10:21:33 -0400; I would like to fit a nonlinear model: y=f(x,b)+error where f is a nonparametric function, such as a loess or gam object, and b is a random effect. I have tried something like f _ loess(t~x) yhat _ n; /archives/html/s-news/2002-08/msg00029.html (7,470 bytes)
19. ___ files (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Tue, 06 Aug 2002 16:34:57 -0400; I have some very large files on my .Data directory (Unix) that start "___" e.g. "___6oe7". Can I safely get rid of these? Naomi S. Altman 814-865-3791 (voice) Associate Professor Dept. of Statistics; /archives/html/s-news/2002-08/msg00041.html (6,629 bytes)
20. Re: ___ files (score: 1): Author: Naomi Altman <naomi@stat.psu.edu>; Date: Wed, 07 Aug 2002 09:56:05 -0400; Thank you to P. Burns, A. Liaw and R. Burbridge who answered. start "___" e.g. "___6oe7". Can I safely get rid of these? The clue, provided by P. Burns, is in S Poetry (see web references below). Her; /archives/html/s-news/2002-08/msg00048.html (8,135 bytes)

Current List: 1 - 20
Page: [1] [2]