EViews.com

In testing my Poisson regression (using the Eviews guide: http://www.eviews.com/help/helpintro.html#page/content%2Flimdep-Count_Models.html%23ww38558) I can see overdispersion, so am switching to a negative binomial distribution. I am comfortable with estimating marginal effects for my Poisson model...

I have followed the below approach to calculating marginal effects, which seems to have worked well for both a binary logit model and a poisson (replacing @dnorm for with @dlogistic for the logit and @exp for the Poisson). eq1.forecast(i) xbf scalar meanxb = @mean(xbf) scalar meffectw = @dnorm(-mean...