* URGENT- POI's built-in quaids command??? Confusion regarding calculation of income elasticity

[sami farook]

Hi all

I am trying to estimate demand and compute income (expenditure elasticity) using QUAIDS as stated in Poi’s Stata Journal “Easy demand-system estimation with quaids, The Stata Journal (2012) 12, Number 3, pp. 433–446”. According to the formula presented here as stated below and the example given (I have run the same example in my Stata (13) using both Poi’s built in stata command and nlsur function evaluator program), the equation is divided by actual expenditure share in the data ‘w’ instead of predicted expenditure share ‘eshare’. Shouldn’t it be divided by predicted expenditure share? Please help.

Income (expenditure) elasticity formula as in Poi’s Journal:

E_i=1+ 1/w_i [β_i+{2λ_i/b(p)* ln(m/a(p))}]

Regards
Sami

[Matt Manierre]

This is something that might be better handled via an email with the author of the command you are using. If you're feeling curious about how the outcome is expected you can always crack open the ado file and see if you can make sense of it. We might be able to tell you what the program does if you find the specific syntax that performs those calculations, but I don't know if many of us will be familiar enough with the substantive question to tell you what it "should" be doing. I certainly am not economist and wouldn't be able to answer that question.