I am using six waves of the PSID to estimate several determinants (particularly wealth) of the wage equation and the selection equation. Stata 16 introduced the xtheckman command, but I can't seem to figure out how to implement weights (either pweights or iweights) in the equation. Could someone please advise? Are weights simply not allowed? In that case, what should/could I do instead?
xtheckman uses random-effects MLE to estimate the coefficients, but I think fixed effects models are more appropriate for estimating wages (confirmed by the Hausman test rejecting the null). Is there a fixed-effects equivalent of xtheckman that can be used to address the selection bias in panel data?
Thank you, in advance, for your help! I hope everyone is safe and healthy (and, alas, productive).
Code:
xtheckman head_wage_rate_LY_log head_male head_age head_age_sq head_edu i.head_race if head_edu!=99 & head_age!=999 & head_race<6 & head_race>0 & head_age>=25 & head_age<=50, select(wtr_head_worked_LY=head_male head_age head_age_sq head_edu head_married num_children_in_FU wealth_w_equity)
Yoy can use xtheckmanfe
ReplyDelete