# RE: st: Bivariate probit model

 From Kristian Jakobsen <[email protected]> To <[email protected]> Subject RE: st: Bivariate probit model Date Sat, 5 Sep 2009 13:07:04 +0000

```Hi again,

Thank you so much for the quick help although I think I have been slightly misunderstood. I am not looking for the sole probability of being poor in period 2 (y2=1), but the  probability of being poor in period 2 conditional on the poverty status in period 1. The easy solution would of course be to run two estimation e.g.:

probit y2 hh if y1=poor
probit y2 hh if y1=non-poor

But I am thinking that there would be a correlation between those two, meaning that the poverty status in period 1 (y1) influences the probability of being poor in period 2 (y2), and this would be confirmed with rho different from 0 in a bivariate probit model. Therefore, I am trying to get a table with the following two columns:

Probability of being poor in period 2 (for poor hhs in period 1) | Probability of being poor in period 2 (for non-poor hhs in period 1)

But how to write that in STATA? As I mentioned previously, I have written:

biprobit y1 y2 hh

but that does not seem right?

Thank you so much for any help,
Kristian
_________________________________________________________________
With Windows Live, you can organize, edit, and share your photos.
http://www.microsoft.com/middleeast/windows/windowslive/products/photo-gallery-edit.aspx
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```