Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: ANOVA Posthocs


From   "Feiveson, Alan H. (JSC-SK311)" <alan.h.feiveson@nasa.gov>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: ANOVA Posthocs
Date   Fri, 22 Oct 2010 15:41:04 -0500

SO this brings to mind the question: 

Is it better (more powerful?) (or more correct?) to do 6 separate FDR analyses with 40 genes each using Scheffe or Tukey adjusted p-values for each gene, or one FDR analysis on 240 unadjusted t-test p-values?

Al Feiveson

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Airey, David C
Sent: Friday, October 22, 2010 2:52 PM
To: statalist@hsphsun2.harvard.edu
Subject: re: st: ANOVA Posthocs

.

Maybe if you are doing 40*6=240 pairwise comparisons, do you want to consider controlling the false discovery rate? Along these lines, if you are doing oneway ANOVAs, why not use -regress- with the -test- command (http://www.stata.com/support/faqs/stat/test1.html) and -post- to file as you run through the genes and then submit the p-values to one of Roger Newson's packages like -smileplot or -qqvalue-?


> Greetings Statalisters.
> 
> I am working on a multiple comparison problem comparing Y by Group with 4 groups, but I have MULTIPLE Y's that I need to consider because my Y's represent fold-increases in gene expression data and I'm working with over 40 genes...
> 
> So in this effort, I'd like to run Oneway ANOVAs of Y by Group for each of my genes, capture the Pairwise comparisons among my groups into a separate dataset, and repeat this with a loop for each of the genes of interest.
> 
> Ideally I'd like to use the Tukey-Kramer Pairwise user-written command (thanks Phil Enders), however that does not provide p-values in the screen output or any matrix in the return list.
> 
> An alternative would be to use the Scheffe posthoc option to -oneway- but I can't figure out how to steal those p-values either?
> 
> Any thoughts on how to grab these p-values?
> Much appreciated...
> Rob
> 


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index