New package on SSC: xtprobitunbal

[Pedro Albarran]

Dear all,

Thanks to Kit Baum, a new package, xtprobitunbal, is now available for download from SSC. In Stata, you can install it by typing

. ssc install xtprobitunbal, all

This package estimates dynamic correlated random effects (CRE) Probit models with unbalanced panels; this command xtprobitunbal estimates these models allowing for the unbalancedness process to be correlated with the time-invariant unobserved heterogeneity. It reduces the computational burden of the maximum likelihood (ML) estimation, while keeping its good asymptotic properties. We also introduce the command mgf_unbal to compute the marginal effects of the variables of the model and its standard errors.

The packages also includes a Stata script and a dataset to provide users with examples that implements these commands

Best wishes,
Pedro

[Aleksandra Anic]

Dear Pedro Albarran ,
Dear other stata users,

I need a help with command xtprobitunbal and mgf_unbal. I get error when I want to calculate any marginal effect. I have several binary variables in the model. I am investigating probability that a child is poor, and as explanatory variables I have age of the child (the only continous variable), and labour market status of parents (1 if one of the parents work 0 if both parents work and 1 if none of the parents work and 0 both parents work). Similarly, I have two variables for education level of parents (1 if education level is medium 0 if educ. level is low and 1 if educ level is high and 0 if educ. level is low).

I get r(2000) no observations. Please find attached the log file.

Additionaly, should I run every time xtprobitunbal command and then mgf_unbal or once when I have run xtprobitunbal I can run several times mgf_unbal (because I need for all variables separately to run mgd_unbal).

Thank you.

Best regards,

Aleksandra