I use GMM and would like, given a set of possible instrumental variables, R to provide me the best combination with lags.
For example
modelo GMM x ~ a b c d
VI q y z
R provide, for example based on one information criterion or another, what is the best combination of q (-1 to x), y , z (-1 to ....)
resulting from the above condition.
I use EViews for this, and the parameters would be similar to that
Equation Specification
Instrument List
Estimation weighting matrix = HAC (Newey-West)
Weight updating = N-step Iterative (number of iterations - for choose)
covariance wieghting matrix = choose between White , HAC, two stages
Linear estimation with 1 weight update
Estimation weighting matrix: HAC (Bartlett kernel, Newey-West fixed
bandwidth = 5.0000)
Standard errors & covariance computed using estimation weighting matrix
Instrument specification: xxxxx(-1 TO -6) yyyyyyyy(-1 TO -9)
Constant added to instrument list