Dear All, I have a problem with 'betareg' function of 'betareg' package, would you, please, help me? I am trying to model compositional similarity on range (0; 1) between pairs of sample units as a function of pairwise geographical distances, and I often get an error message like this:
"Error in chol.default(K) : the leading minor of order 3 is not positive definite In addition: Warning message: In sqrt(wpp) : NaNs produced Error in chol.default(K) : the leading minor of order 3 is not positive definite In addition: Warning messages: 1: In betareg.fit(X, Y, Z, weights, offset, link, link.phi, type, control) : failed to invert the information matrix: iteration stopped prematurely 2: In sqrt(wpp) : NaNs produced" The problem occurs with any link function. I read that this error is caused when my matrix is not positive definite. But how should I avoid this if these are the data I have? Thank you and best regards, Attila Lengyel [[alternative HTML version deleted]] _______________________________________________ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology