Dear forum members,
I would ask you a help on two issues please.
(a) I would like to run a two stage hedonic price model which involes use a coefficient estimated in a 1st stage of regression as independent variable in a 2nd stage regression. The 1st equation is yit= pit+ zi+eit, where y=quantity of good sold, p=price, z= a vector relevant product attributes and eit is an idiosyncratic error term.
The 2nd stage equation is (coeff.p)= zit+vit.
Question 1: how I can write (coeff.p) is Stata language?
(b) Second issue is related to characteristics of z variables. These are categorical variables with 3-5 levels of coding for different attributes.
Questionn 2: how I set a reference base for each variables estimate their coefficients against the base reference, whilst also avoiding the dummy (or in this case indicator) trap problem in a fixed effects model used?
Thanks,
Mohamud
I would ask you a help on two issues please.
(a) I would like to run a two stage hedonic price model which involes use a coefficient estimated in a 1st stage of regression as independent variable in a 2nd stage regression. The 1st equation is yit= pit+ zi+eit, where y=quantity of good sold, p=price, z= a vector relevant product attributes and eit is an idiosyncratic error term.
The 2nd stage equation is (coeff.p)= zit+vit.
Question 1: how I can write (coeff.p) is Stata language?
(b) Second issue is related to characteristics of z variables. These are categorical variables with 3-5 levels of coding for different attributes.
Questionn 2: how I set a reference base for each variables estimate their coefficients against the base reference, whilst also avoiding the dummy (or in this case indicator) trap problem in a fixed effects model used?
Thanks,
Mohamud
