Exclusion of cross-sections in GLM

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

Sebastian
Posts: 3
Joined: Tue Jun 22, 2021 1:42 pm

Exclusion of cross-sections in GLM

Postby Sebastian » Thu Jun 24, 2021 1:31 pm

I am analyzing a dataset consisting of 119 countries from 1985 to 2015 using generalized linear models and I have two questions.

First: Does EViews drop cross-sections with too much missing values in GLM? When I run an OLS (panel least squares) with the same data, EViews reports that only 107 cross-sections have been included. For GLM the number of cross-sections are not reported in the estimation results. However, it would be useful to know if EViews does also exclude countries in GLM.

Second: If cross-sections are excluded, is there a way to find out, which ones? Of course, it would be important to know which countries are included and excluded in the analysis.

Hoping the questions are not too stupid, but I did not find an answer in the User’s Guide.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13306
Joined: Tue Sep 16, 2008 5:38 pm

Re: Exclusion of cross-sections in GLM

Postby EViews Gareth » Fri Jun 25, 2021 8:45 am

Both least squares and GLM will drop observations that are NAs. If a cross-section has all NAs, then it will be completely excluded. Least squares is a bit smarter in reporting how many were dropped, but that's just a reporting issue - both should drop the same number.
Follow us on Twitter @IHSEViews


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 36 guests