Hello,
I would like to measure the standard error of the wage-elasticity of labor supply which is a function of wage-elasticity of separation to employment and to non-employment.
First, I estimate the wage-elasticity of separation to employment by performing a logit model on the probability of separation to employment, the elasticity is then obtained from the "margins dyex" command.
Second, I estimate the wage-elasticity of separation to non-employment by performing a logit model on the probability of separation to non-employment, the elasticity is then obtained from the "margins dyex" command.
The idea is then to store the estimates so that I can run "lincom" command to measure the wage-elasticity of labor supply. I can store the estimates but I get an error message when I try to "suest" the estimates. Since the SE of wage-elasticity of separation are obtianed using the delta method, I cannot use the suest command. I am wondering whether there is another command that allows for nonstandard VCE?
Thanks!
I would like to measure the standard error of the wage-elasticity of labor supply which is a function of wage-elasticity of separation to employment and to non-employment.
First, I estimate the wage-elasticity of separation to employment by performing a logit model on the probability of separation to employment, the elasticity is then obtained from the "margins dyex" command.
Second, I estimate the wage-elasticity of separation to non-employment by performing a logit model on the probability of separation to non-employment, the elasticity is then obtained from the "margins dyex" command.
The idea is then to store the estimates so that I can run "lincom" command to measure the wage-elasticity of labor supply. I can store the estimates but I get an error message when I try to "suest" the estimates. Since the SE of wage-elasticity of separation are obtianed using the delta method, I cannot use the suest command. I am wondering whether there is another command that allows for nonstandard VCE?
Thanks!
Comment