Dear Statalisters,
I am using a neighborhood dataset for Bogotá, Colombia and i am using the user written command spreg to estimate a GS2SLS for a cross section dataset of urban crime.
i can calculate direct, indirect and total effects using the postestimate commands from spreg (2011, Drukker et al.), but to the best of my knowledge is not directly possible to do any kind of inference for these effects using the spreg command or any command from the sppack.
Could anybody point me where or how can i calculate measures of dispersion for the effect estimates after using the spreg command? Lesage and Pace (2009) suggest using mcmc but i dont really know how to do these draws in stata.
Thank you
Daniel Martinez
I am using a neighborhood dataset for Bogotá, Colombia and i am using the user written command spreg to estimate a GS2SLS for a cross section dataset of urban crime.
i can calculate direct, indirect and total effects using the postestimate commands from spreg (2011, Drukker et al.), but to the best of my knowledge is not directly possible to do any kind of inference for these effects using the spreg command or any command from the sppack.
Could anybody point me where or how can i calculate measures of dispersion for the effect estimates after using the spreg command? Lesage and Pace (2009) suggest using mcmc but i dont really know how to do these draws in stata.
Thank you
Daniel Martinez
