Hi everyone, I have a time series data where there are several variables given.For example, disappearance of beef, pork, turkey and chicken and their retail prices (retail prices of beef...chicken), their cpi (cpi_beef...cpi_chicken), population, disposable income and overall CPI. I calculated expenditure share by the following command:

gen expmeat= cpiturkey+ cpichicken+ cpibeef+ cpiporkfresh
gen wbeef= cpibeef/ expmeat
gen wpork= cpiporkfresh / expmeat
gen wchicken= cpichicken / expmeat
gen wturkey= cpiturkey / expmeat

Is it alright?

Now, I am trying to estimate LA-AIDS model in stata. So far, I have got two commands: aidsills and quaids from internet sources but none of them provide me with how to calculate LA-AIDS (I might be wrong and missing something here). My questions are:

1. How do I calculate the stone price index which is required for LA-AIDS? I calculated expenditure share (wi), but I am not sure how to calculate stone price index from these data. I know it is lnp= sum(wi*lnpi), so here we have got wi, but what is lnpi? should it be retail price or CPI? For example, stone price index=w(beef)*cpi_beef+...+w(chicken)*cpi_chicken or is it stone price index=w(beef)*retail_beef+...+w(chicken)*retail_ch icken?

2. What is the command for LA-AIDS in stata? I think in general AIDS, we use nlsur but for LA-AIDS, we use seemingly unrelated regression (SUR). But I am not sure how to run LAAIDS. For example, how do I fit homogeneity, symmetry conditions in SUR? in aidsills command and qaids, we can impose homogeneity, symmetry conditions with the commands themselves. But not sure how to do it with SUR?