I am trying to calculate marginal effect using negative binomial regression outputs. I know there is command to calculate marginal effect in STATA, R, SAS but I used SPSS to to negative binomial regression and don't have access to other statistical software due to limited access to data and software. So, I am wondering if there is any way to calculate ME using regression outputs? I would be grateful to you if somebody can suggest me the proper method.
Update: I figured out that conditional mean (also called as lamda)*B-coefficient gives marginal effect in the Poisson regression model. It could be helpful to others as well, So I wanna share.
More Links:
http://support.sas.com/documentation/cdl/en/etsug/68148/HTML/default/viewer.htm#etsug_countreg_details05.htm
http://support.sas.com/documentation/cdl/en/etsug/68148/HTML/default/viewer.htm#etsug_countreg_details05.htm
If in case, you are using SPSS or software without the tool to calculate Marginal Effect, then only way is to calculate marginal effect is to calculate "conditional mean" (not the absolute mean) using Excel. The process is explained in the link below.
http://www.techrepublic.com/blog/windows-and-office/two-ways-to-find-a-conditional-average-in-excel/
http://support.sas.com/documentation/cdl/en/etsug/68148/HTML/default/viewer.htm#etsug_countreg_details05.htm
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS