How I can test if price elasticity in a quantile regression differs among quantiles?

[johny kos]

Dear statalists,

I would like to ask you how I could test if price elasticity differs statisticall ysignificantly between quantiles. For example after running

xtqreg lnq lnp, i(id) q(0.25)

and then

xtqreg lnq lnp, i(id) q(0.50)

and

xtqreg lnq lnp, i(id) q(0.75)

, could I test if price elasticity coefficients differ?
Thanks a lot in advance.

[Oscar Ozfidan]

Johny, I dont know the answer to your question but I will try to help by pointing out to others that what you are asking (for those who are not familiar with what elasticity is) is simply a comparison of the coefficient of lnp for different quantiles. Hopefully this will help you get some responses....

[Joao Santos Silva]

Dear johny kos,

There are several ways of doing that, some will require more work than others. A simple way of doing it is to do a bootstrap for the difference and see if that is significant. A simpler way of getting some information about your question is simply to use the ls option and see if the scale coefficient on lnp is significant.

Best wishes,

Joao

[johny kos]

Dear Oscar and Joao,
I would like to thank you for your response. Exactly, Oscar, this is my question. Thank you a lot.
Regarding the bootstraping, is there any command or code that I could read or run for finding if the difference between percentiles is statistically different?
Thank you very much in advance.