Hello everyone.
I am running an analysis using Stata 17 and the command xtabond2, ivstyle(all covariates + external instrument). The dependent variable is expressed in log.
I would like to quantify the results in terms of a standard deviation change of my variable of interest (similar to what I would do with standardised coefficients). However, due to the fact that this is a two-stage GMM estimations I am not sure how I should proceed. I tried both to standardise the coefficients myself and calculate the margins (atmeans) of a one standard deviation change of my variable of interest, but the results do not look quite correct and I suspect that I am missing something.
Do you have any solution/suggestion?
Thanks in advance
Valentina
I am running an analysis using Stata 17 and the command xtabond2, ivstyle(all covariates + external instrument). The dependent variable is expressed in log.
I would like to quantify the results in terms of a standard deviation change of my variable of interest (similar to what I would do with standardised coefficients). However, due to the fact that this is a two-stage GMM estimations I am not sure how I should proceed. I tried both to standardise the coefficients myself and calculate the margins (atmeans) of a one standard deviation change of my variable of interest, but the results do not look quite correct and I suspect that I am missing something.
Do you have any solution/suggestion?
Thanks in advance
Valentina
Comment