Hi all,
need help very urgently

I did stepwise logistic regression for 35 covariates and added one SNP out
of (500000) to get the best model for each model
As my professor asked me

using this command,


outfiles <- paste(colnames(snps), ".txt", sep="") # list of output files
for the best models
for(i in 1:ncol(snps)) {
  model <- glm (Pheno~var1+var2+var3+..(all covariates)...+snps[,i] ,
data=covar , family=binomial("logit"))
  st <- step(model)
  out <- data.frame(FID=covar$FID, IID=covar$IID, st$model) # formatting
for plink
  write.table(out, outfiles[i], col.names=TRUE, row.names=FALSE,
quote=FALSE, sep="\t")
}

and i end up having 500,000 files for each SNP

1- Is this way is approperiate in assessing the fittness of logistic model ?
2- how can i now use all of these file in a way that i dont need to write
the glm formula for each file ?
3- Is there is a better way to save the results so i will not have to deal
with 500,000 files ?

Please need your help


Alajmi

        [[alternative HTML version deleted]]

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to