[R] logit problem

yoo hoo freesuccess2001 at yahoo.com
Fri Jun 22 08:46:01 CEST 2007


Hi there,
 
 I was trying to fit this dataset into LR model. This dataset includes 18 normal and 17 
 cancer. There are totally 14 markers (7 mRNAs and 7 Proteins). When I fitted into LR 
 model, R gave me warning:
 Warning messages:
 1: algorithm did not converge in: glm.fit(x = X, y = Y, weights = weights, start = start, 
 etastart = etastart,
 2: fitted probabilities numerically 0 or 1 occurred in: glm.fit(x = X, y = Y, weights = 
 weights, start = start, etastart = etastart,
 
 I don't know why the algorithm did not converge. Is it because sample size/marker is 
 small? Is it possible you run this dataset in s-plus? Thank you very much for your help.
 
 Michael
       
---------------------------------



More information about the R-help mailing list