multiple logistic regression
multinom() in package nnet.
On Fri, 26 Nov 2004, Anne Piotet wrote:
Is there an utility for multiple logistic regression where the response is not ordinal? As the predictors do not satisfy the necessary hypothesis, I cannot use discriminant analysis.
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