Multinomial logistic regression under R and Stata
On Mon, 21 Apr 2003, Tak Wing Chan wrote:
the Hessian when the scale of a variable is very small. I talked to a colleague, David Firth, about this, and he suggests
> Possibly it would be worth implementing an algebraic vcov method for
multinomial logit models [in R]?
Yes, and I look forward to receiving your contribution of it. (See the R startup banner.)
Brian D. Ripley, ripley at stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UK Fax: +44 1865 272595