I'm currently using the leasqr function among a large set of data (function is called up to 310 times), and i sometimes recover NaN values for parameters covariance & correlation matrices (covp and corp). Is there a way to fix this up ?
I may add i also recover sometimes non-symetrics parameters covariance matrices.
Thanks for replies; Regards,
Grostuff
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail