Hi,
My probit model will not achieve convergence with the addition of my final variable of interest, despite this variable having a significant relationship with the outcome in univariate analysis and no strong correlations with any other covariates. I have 8 total covariates with inclusion of the final variable. When I run my two part model as two individual probit models, with the requirement of the selection variable equalling "1" for the second step model, both models achieved convergence. Does anyone have any insight into why this may be?
Thanks!
My probit model will not achieve convergence with the addition of my final variable of interest, despite this variable having a significant relationship with the outcome in univariate analysis and no strong correlations with any other covariates. I have 8 total covariates with inclusion of the final variable. When I run my two part model as two individual probit models, with the requirement of the selection variable equalling "1" for the second step model, both models achieved convergence. Does anyone have any insight into why this may be?
Thanks!
