Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Poisson and marginal effects

From   Jeffrey Wooldridge <>
Subject   Re: st: Poisson and marginal effects
Date   Sun, 13 May 2012 18:41:13 -0400

In Poisson regression the average partial effect of a continuous
variable is just the sample average of y times the coefficient.
Naturally, it is harder for a discrete change. There is makes sense to
obtain the predicted value at the two different settings of the
explanatory variable, with other variables evaluated at their observed
values, and average the difference.

The reason this is not regularly done in Poisson regression is that
the coefficients have interpretations as percentage changes.

On Sun, May 13, 2012 at 5:29 PM, Jessie C <> wrote:
> If I am understanding Poisson regression correctly, its interpretation
> is based on evaluating the coefficients at a certain data point like
> probit.  In probit, there is a marginal effects model, mfx, is that
> right?  That evaluates the probit results at the mean of the data.  Is
> there an analogue for poisson regression?
> *
> *   For searches and help try:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index