[R] logistic regression with temporal correlation

[Prof Brian Ripley]

This is not `temporal correlation'.  You can supply a weights argument to 
glm and so downweight older observations.  Since the theory is no longer 
exactly apposite, treat the standard errors etc with some caution.

On Thu, 18 Mar 2004 merser at tiscali.dk wrote:

> We would like to perform a logistic regression analysis weighting the
> independent variable in  a temporal fashion, i.e. events occuring most
> recent get highest weight. Does anyone know how to do this in R??
> 
> Regards
> S. Merser and S. Lophaven

-- 
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