So I have a system of equations made up with some time series, to be estimated with a Generalised Moments Method model. Sth like:
PREM[t] = phi_0+phi_1*PREM[t-1]+phi_2*IR[t-1]+phi_3*INAD[t-1]+phi_4*U[t-1]+phi_5*RR[t-1]+e[t] SPREAD[t]=alpha_0+alpha_1*SPREAD[t-1]+alpha_2*IR[t-1]+alpha_3*PREM[t-1]+alpha_4*RR[t-1]+alpha_5*EMBI[t-1]+u[t]
where each variable is a time series, t-i is current period minus i periods, e and u are both white noise.
The model to be estimated is a GMM with a HAC covariance matrix. I want to estimate this model on the data I have, but I have no idea how to do that in R, I only know how to do that on EViews ( http://www.eviews.com/help/helpintro.html#page/content%2Fsystem-How_to_Create_and_Specify_a_System.html%23ww37136 exactly as they do on the part of "GMM - Time Series (HAC)").
However, I neither know how to do it on R or have EViews disponible on my PC. How do I do this on R?
Also, let's say that I don't know the best fit for lags of each variable on my system of equations, wanting to select the model based on an information criteria, like AIC or BIC. How to do this with this case? Is there any pre-sets that already do that?
