Re: Non-positive semi-definite message

From: "bvatul" <bvatul@ufl.edu> Subject: Re: Non-positive semi-definite message Date: Wed, 15 Aug 2001 12:36:27 -0400 Hello Alan I had a similar error message when I was analysing a data set with three = compartment model for drug and a similar model for metabolite with $COV = step. The problem could be solved (Thanks to Dr Peter Bonate) by viewing = the scatter plots of various parameters and minimising them in the model = if they are correlated and changing the NSIG from default of 3 to 5. = One more suggestion was to use MATRIX=3DS in COV step to get around this = message although I am not sure how it works.=20 I used these methods and I could work around the problem. It is not = clear to me still how change of NSIG effects the search of the global = minima? Perhaps somebody could comment on this. Hope this helps Atul