Journal of Computations & Modelling

Econometric Computing Issues with Logit Regression Models: The Case of Observation-Specific and Group Dummy Variables

• [ Download ]
• Times downloaded: 11483

• Abstract

  Our study extends work on econometric computing issues in logit regression models by focusing on observation-specific and group dummy variables, wherein all or nearly all of the members of the group are associated with the same value for y, rather than the case of continuous regressors. To make our case, we employ a small data set from a previously published study. Lastly, we explore, using various econometric software packages, several prescriptions for dealing with these issues.