Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Mirko Moro <mirko.moro@gmail.com> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: indirect effects of indep vars in selection model on dep var |

Date |
Tue, 24 Jan 2012 14:16:40 +0100 |

Dear Statalisters, I am interested in estimating the effect of a variable Z on Y, where Y can be either an ordered or a continuous variable. I know from my theory that Z has an indirect effect on Y via a direct effect on W, where W is a dummy variable. I also know that W and Z depends on common set of factors F. So I am thinking of estimating a system such as: Y = Wa + Fb + e Prob(W) = Zc + Fd + u. I have cross-sectional data and I am thinking to use -treatreg- here when Y is a continuos and or -cmp- by Roodman if Y is an ordered response variable. First, is this correct? Second, one of my first research question is how much (the probability of) Y changes because of Z via its indirect effect on W? Is it possible to get an idea of the size of the indirect association? A second research question linked to this would be how much the vector F affects Y only via W? In other words, I am not that interested in W itself...but in the size of the indirect effects of the variables in the "selection model" . Any suggestion on how these sizes could be reported and interpreted? Best regards, Mirko -- Mirko Moro Lecturer in Economics Economics Division University of Stirling FK94LA Scotland (UK) t: +44(0)1786467479 f: +44(0)1786467469 e: mirko.moro@stir.ac.uk * * 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/

- Prev by Date:
**st: left-side censoring of the independent variable** - Next by Date:
**RE: st: scientific notation turn off** - Previous by thread:
**st: left-side censoring of the independent variable** - Next by thread:
**st: svyset and complex design** - Index(es):