Re: [R] Quantile Regression Question

2007-11-14 Thread roger koenker
Package questions to package maintainers please. "non-positive fis" is not an error, it is a warning -- if the number of negative fis is large relative to the sample size then there is some reason to doubt the plausibility of the specification of the model specified by the rq formula.

[R] Quantile Regression Question

2007-11-14 Thread Arti Mann
Hi, Could you please explain what is non-positive fis error? I have been trying to use quantile regression (rq) procedure and I keep ending up with this error. I haven't been able to find an explanation for the same. Best Regards, Arti Arti Mann Ph.D. Student Department of Information Systems