[R] Lapack Routine Error

Shaibu Pompo opomporay at yahoo.com
Fri Jul 13 19:38:12 CEST 2007


I am running a simulation study to assess the type I error rate of a likelihood ratio test. This requires me to minimize over restricted spaces under the null and under the alternative hypothesis. My problem is that occassionally I run into an error message that reads: "Error in solve.default(cov, ...) : Lapack routine dgesv: system is exactly singular" when I make a call to "nlminb". The negative log likelihood function being minimized is based on the multivariate normal distribution. Can anyone help me with the source of such an error message? Or better still how to get at the source?

SP: opomporay at yahoo.com

Got a little couch potato? 
Check out fun summer activities for kids.

More information about the R-help mailing list