Hello,
I am thinking about the best way to estimate a spatial lag model with a time dimension in STATA. I am considering a 2SLS IV estimation where the instruments are constructed from the independent variables multiplied by the weighting matrix. Another possibility would be to use the spatreg command by Maurizio Pisati. I know that this set of commands is usually used on cross-sections only. But I am wondering what speaks against construction the weighting matrix in a way that it accounts for the time dimension. For instance, if W1 denotes the weighting matrix for year 1 and Wn the weighting matrix for the last year in the sample, than one could construct the following weighting matrix:
W1_0__0
0__.__0
0__0__Wn
This weighting matrix could then be multiplied with the vector of the dependent variable. As far as I understand, such an approach is applied in the paper stated below.
Bruce A. Blonigen1, Ronald B. Davies_, Glen R. Waddell, Helen T. Naughton, FDI in space: Spatial autoregressive relationships in foreign direct investment, European Economic Review 51 (2007) 1303â??1325.
Would such an approach be okay?
Best regards
Phil
--
Psssst! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger01
*
* 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/
