Dear all,
Re: Bug in StataIC/10.1
There seems to be a bug in Stata 10. I attached the dataset that I used to fir a Cox PH model. It is simply a survival-time dataset which has already been stset. The covariate is x and I try to fit a univariate Cox PH model with x as the sole covariate. So I typed:
stcox x
But Stata 10 gives me an error massage as:
"flat region resulting in a missing likelihood"
Please brear in mind that I can perform the same analysis in Stata 9 without any problem. I would be grateful if someone can give me a clue on what is causing this error massege in Stata 10 but NOT in Stata 9.
Looking forward to hearing from you.
Babak Oskooei
MRC Clinical Trials Unit
London
Attachment:
Data_that_reveals_bug.dta
Description: Binary data
