Hi Stata users,
I am setting up a distributed lag model in xtabond, and I am a bit puzzled about what xtabond is doing exactly when I restrict the instrument set of the distributed lags.
Say I want to model:
y(it) = y(i t-1) + x(it) + x(i t-1) + e(it)
where x is a predetermined variable. Suppose I want to restrict the instrumens of x to just one further lag of x. There are two ways to write the command, and they should be identical:
1. xtabond y , lags(1) pre(x, lag(1,1) )
2. xtabond y , lags(1) pre(x lag_x, lag(0,1) )
where lag_x is a variable I created by typing:
generate lag_x = L.x
These two commands give different results, which is puzzling. On the other hand, 2. gives, strangely enough, identical results to the following command:
3. xtabond y , lags(1) pre(x, lag(1,2) )
My question is: If I want to restrict the instruments of x to just one lag, which of the commands above should I use? Asked more open-endedly: What is it that 1. is doing versus that which 2. and 3. are doing in terms of the instrument set they are setting up?
Thanks for any thoughts on this.
Tewodaj
~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
Tewodaj Mogues
Dept. of Agricultural and Applied Economics
University of Wisconsin - Madison
427 Lorch St. #317, Taylor Hall
Madison, WI 53706
*
* 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/