Forum: help

RE: micEconAids [ Reply ] By: Arne Henningsen on 2017-06-22 18:29	[forum:45203]
You can use arguments "instNames" and "shifterNames" of aidsEst() to specify instrumental variables and demand shifters, respectively. You can use both of these arguments at the same time. In this case, you should include all exogenous demand shifters also in "shifterNames".

micEconAids [ Reply ]
By: miguel henry on 2017-06-22 13:13

[forum:45202]

Hi Arne,

In the process of estimating the AIDS model, I am interested in allowing for two extra features: demand shifters (e.g., demographic variables) and endogenous regressors (e.g., endogenous prices and endogenous total expenditure).

For demand shifters the most common approach is to use the translating approach of Pollak and Wales (1981), which I think you refer in some of your documentation (based on the equation) as the adjusted alpha_i coefficients.

For endogenous regressors, we can introduce IVs (e.g., IVs for the prices and IVs for total expenditure) and augmented regression techniques from Hausman (1978) and Holly and Sargan (1982).

My question to you: Since I do not see any implementation allowing for both demand shifters and endogenous regressors using your R package micEconAids, is this currently feasible? And if so, could you please provide some useful directions?

This will be very useful since it is a need in current empirical applications, and as far as I know there is only one software package that offers this possibility, but first it is not implemented in the R environment (it is a Stata package) and second it does not offer many of great options that micEconAids does.

Thank you,
Miguel