SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 

 View Only
Back to discussions
Expand all | Collapse all

Converting logit regression coefficient for interaction to marginal effect in GEE model

  • 1.  Converting logit regression coefficient for interaction to marginal effect in GEE model

    Posted Fri September 08, 2023 10:28 AM

    As in the usual logit regression model, is the conversion of B to marginal effects (percentage points) the same in the GEE model with subject clustering and exchangeable working correlation plus many covariates?

    For example, a logit regression B coefficient of .68 in GEE would convert to roughly 13 percentage points if the Y proportion is .26 (.26 X (1- .26) x .68

    A separate issue is that this coefficient does not match the marginal means estimates of the 2 by 2 interaction. I would gather the interaction coefficient is the correct estimate.



    ------------------------------
    tille johns
    ------------------------------



Related Content