Confidence Intervals for Function of Estimated Parameters
Hi everyone,
I am using Stata18 and I have derived from a demand equation with linear and quadratic prices the following price elasticity, η_t:
η_t = α_1 + 2 * α_2 * price_t
where price_t is the time-t measure of the price, and α_1, α_2 are the estimated coefficients associated to the linear and quadratic price terms. Estimation of these two coefficients is carried out with Newey-West standard errors using the newey command with lags(3) and there are 146 observations in the dataset.
I would like to compute a 95% confidence interval for my η_t. I was thinking about bootstrapping, but I really don't know how to do this given that η_t is time varying in price_t, and thus I cannot run:
bootstrap eta_t = (_b[price1] + 2 *_b[price2] * price_t), ...
Any help would be appreciated. Thank you very much in advance!
I am using Stata18 and I have derived from a demand equation with linear and quadratic prices the following price elasticity, η_t:
η_t = α_1 + 2 * α_2 * price_t
where price_t is the time-t measure of the price, and α_1, α_2 are the estimated coefficients associated to the linear and quadratic price terms. Estimation of these two coefficients is carried out with Newey-West standard errors using the newey command with lags(3) and there are 146 observations in the dataset.
I would like to compute a 95% confidence interval for my η_t. I was thinking about bootstrapping, but I really don't know how to do this given that η_t is time varying in price_t, and thus I cannot run:
bootstrap eta_t = (_b[price1] + 2 *_b[price2] * price_t), ...
Any help would be appreciated. Thank you very much in advance!
Comment