[R] FWER + multiple linear models

Serguei Baranov barano_s at molgen.mpg.de
Fri Jul 30 17:41:05 CEST 2004


Could someone kindly help me with the following question:
when I analyze microarray data I need to fit multiple linear regression 
models between genes and clinical patameters followed by estimation of 
the p-values.
What's the solution to implement Westfall and Young's algorithm + 
resampling into the scheme:
lm -> stepAIC -> anova.
Actually permcor works fine for me in the case of one regression 
variable, I need extension for multiple regression models.
Any ideas are very much wellcome!

Sincerely yours,
Serguei Baranov
Research "Obesity & Hypertension" MPI of Molecular Genetics, Ihnestr. 
73, 14195, Berlin, Germany
phone:   +49 30 84131238
fax:        +49 30 84131228




More information about the R-help mailing list