Search String: Display: Description: Sort:

Results:

References: [ +from:huangp@musc.edu: 21 ]

Total 21 documents matching your query.

1. loop over a series regression models (score: 1)
Author: .net>
Date: Thu, 14 Apr 2005 23:27:28 -0400
tion = function(a
/archives/html/s-news/2005-04/msg00107.html (8,422 bytes)

2. Re: loop over a series regression models (score: 1)
Author: .com>
Date: Fri, 15 Apr 2005 11:28:11 -0400
or (n=12) sampled
/archives/html/s-news/2005-04/msg00114.html (15,017 bytes)

3. Re: loop over a series regression models (score: 1)
Author: .edu>
Date: Fri, 15 Apr 2005 11:34:26 -0400
ication) with the
/archives/html/s-news/2005-04/msg00115.html (15,751 bytes)

4. Re: loop over a series regression models (score: 1)
Author: o.au>
Date: Sat, 16 Apr 2005 17:39:35 -0400
nce I want to per
/archives/html/s-news/2005-04/msg00121.html (12,124 bytes)

5. How to force a for loop continue even there are errors? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 12 Oct 2004 00:00:33 -0400
Dear S list members, I have a dataframe called mydata with column names c("y1", ...,"y400","group","male","age","Education.Level") I wrote a loop to run a series of regressions with response variable
/archives/html/s-news/2004-10/msg00037.html (7,975 bytes)

6. Re: How to force a for loop continue even there are errors? [summary] (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 12 Oct 2004 13:41:22 -0400
I want to thank the helpful suggestions from Max Zhao, Bill Dunlap, Dimitris Rizopoulos and Shawn Boles. Max suggested to use option "singular.ok=T" to deal with collinearity. Bill suggested to use x
/archives/html/s-news/2004-10/msg00042.html (9,332 bytes)

7. Re: Multiple R-Squared (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Mon, 05 Apr 2004 14:25:31 -0400
I guess that R^2 from the intercept model is always smaller than the R^2 from the non-intercept model. Below is what I thought. Please correct me if I misunderstood anything.. Let SSR = regression s
/archives/html/s-news/2004-04/msg00018.html (9,395 bytes)

8. Re: How to find the dates? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Mon, 01 Mar 2004 11:32:31 -0500
The event is actually "the present of a symptom". Patient's visits are not equally spaced. We are looking at the first time the patient has the symptom, and how long it takes the symptom will disapp
/archives/html/s-news/2004-03/msg00007.html (9,226 bytes)

9. Summary: How to find the dates? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 02 Mar 2004 13:52:48 -0500
Many thanks to Dimitris Rizopoulos and Sam Buttrey who suggested the use of split and lapply (or sapply). My original question is: I have a very large cohort data that contains repeated measures. The
/archives/html/s-news/2004-03/msg00024.html (13,981 bytes)

10. Residuals from logistic regression? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 09 Mar 2004 12:53:12 -0500
Dear All, When I fit a logistic regression, say, using out<- glm(y ~ x1 + x2, family = binomial(link = logit)) How can I find what type of residuals that out$residuals gives me? How can I check (what
/archives/html/s-news/2004-03/msg00095.html (6,950 bytes)

11. Re: Residuals from logistic regression? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 09 Mar 2004 14:49:02 -0500
Many thanks to professor Brian Ripley, Sundar Dorai-Raj and Peter Alspach for their quick responses to my question. The suggestions are very helpful. Thanks again! Regards, Peng Huang Dear All, When
/archives/html/s-news/2004-03/msg00099.html (9,392 bytes)

12. How to find the dates? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Sun, 29 Feb 2004 17:41:39 -0500
Dear All, I have a very large cohort data that contains repeated measures. The data looks like: id days event 1 0 0 1 30 0 1 50 0 1 95 1 1 120 1 1 147 0 1 180 0 2 0 0 2 75 0 2 100 0 3 0 0 3 37 1 3 80
/archives/html/s-news/2004-02/msg00226.html (7,511 bytes)

13. reference on rank based model? (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Fri, 05 Dec 2003 11:55:50 -0500
Dear All, Could some one point out to me some references on rank based models or tests for multivariate outcomes measured longitudinally? All outcomes are correlated and repeated measures are depende
/archives/html/s-news/2003-12/msg00022.html (7,063 bytes)

14. test for interaction (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Fri, 14 Nov 2003 14:45:16 -0500
.
/archives/html/s-news/2003-11/msg00101.html (8,192 bytes)

15. Re: test for interaction (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Fri, 14 Nov 2003 19:39:55 -0500
i
/archives/html/s-news/2003-11/msg00103.html (10,361 bytes)

16. Re: test for interaction (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Fri, 14 Nov 2003 21:44:51 -0500
2
/archives/html/s-news/2003-11/msg00106.html (10,884 bytes)

17. Re: test for interaction (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Fri, 14 Nov 2003 22:05:41 -0500
s
/archives/html/s-news/2003-11/msg00107.html (12,319 bytes)

18. Re: Usint testvariables for further calculations (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Wed, 04 Jun 2003 11:53:48 -0400
x _ 2*pval$p.value -- Peng Huang, Ph.D. Assistant Professor Department of Biometry and Epidemiology Medical University of South Carolina 135 Cannon Street, Suite 303 Post Office Box 250835 Charlesto
/archives/html/s-news/2003-06/msg00024.html (8,761 bytes)

19. Re: lag down (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Mon, 16 Jun 2003 13:45:25 -0400
If there are more than two observations per unique id, we may try the following: id x 1 2 33 2 2 NA 3 3 67 4 3 NA 5 3 NA 6 3 NA 7 3 NA 8 4 24 9 4 NA 10 4 NA 11 4 NA 12 5 45 13 5 NA 14 5 NA 15 5 NA 16
/archives/html/s-news/2003-06/msg00099.html (8,905 bytes)

20. Re: summary lag down (score: 1)
Author: Peng Huang <huangp@musc.edu>
Date: Tue, 17 Jun 2003 10:26:33 -0400
I guess the reason you got error message from my code is because the code can not be run for the second time. Here is why. Let's create a data set: id <- sample(1:3,10,T) id<- sort(id) x1<- rle(id)
/archives/html/s-news/2003-06/msg00115.html (12,575 bytes)


This search system is powered by Namazu