Re: st: Micro and Macro Variables in Panel Data

[Andreas Kuhn <akuhn@access.unizh.ch>]

REINERTSON,M YVONNE wrote:
> Dear Statalisters,
>
> I have gained greatly by reading this list over the years while 
> completing my college core courses (possibly more than I gained from my 
> various stat classes). However, now that I am an active researcher, and 
> devoted Stata follower, I need some guidance. In addition, someday, 
> after completing enough of a rather steep learning curve, return the 
> favor by contributing to the list.
>
> Problem to Solve: I am working on research that involves a panel data 
> where the LHS response variable is an individual specific microeconomic 
> variable while the RHS explanatory variables are all macroeconomic 
> variables common to each panel. Obviously, Stata is not able to break 
> out the relationships with this panel data structure.
>
> Solutions Attempted: I am currently studying chapter eleven in Kmenta, 
> 1986, “Elements of Econometrics” to see if I can use an indicator 
> variable specific to each individual in an interaction term with the 
> macroeconomic variables. Something like Y(i,t)=B(0) + B(1)I(i)X(t) 
> +…+e(i,t)
>
> Questions: (1) Statalist Archives: There have been previous threads on 
> Statalist concerning this research problem, but I have not been able to 
> search out any useful past postings. Does anyone remember this topic 
> being discussed? If so, do you remember the Stata solutions?; (2) 
> Literature Research: What are some literature references that I could 
> research on this topic?
>
> My field of study is economics; however, the references that you 
> recommend need not be. I’m really looking for technique guidance, which 
> can come from any field of research.
>
> Version: Stata 7 (final update)
> OS: WindowsXP
>
> Thanks,
>
> Yvonne
>
> M. Yvonne Reinertson
> University of Florida, PhD Candidate
> myvonne@ufl.edu
>
>
> --
> REINERTSON,M YVONNE
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
Dear Yvonne,

I think you should have a look at so-called multilevel / hierarchical / 
random coefficients - models, which try (among other things) to tackle 
exactly to problem you face: explaining mirco-data (partially) through 
macro-data.

I am not very knowledgeable about these methods, but I just wanted to 
give this hint. You can obatin more information at, for example:

www.gllamm.org

where you can also donwload stata-programs - gllamm - for estimating 
these models. An easy-going introduction is given by: Thomas DiPrete and 
Jerry Forristal (1994), Annual Review of Sociology, 20, 331-357.

I hope this helps, Andreas

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/