In an earnings regression (log linear, i.e. coefficient is an approximate % change), with gender why doesn't the coefficient value change (except the sign which obviously reverses) when you alternate between male and female as the base reference category? Gender pay gap % diff should depend on base?
Related Posts with Swapping male/female as base category in earnings regression model
Spatial Correlation in Standard ErrorsHi all, I would like to implement Conley standard errors to allow for spatial correlation, but I'm n…
Interpreting interaction term coefficient with PPMLDear all, I am trying to estimate an interaction term between two continuous variables in a PPML mo…
Insignificant direct effect and significant indirect effect with -ldecomp- command (Mediating analysis)Hello! My name is Josep Maria Caroz and I am doing a research paper about the relationship between e…
Generating the average of the last 3 observed values of a variable in a panel dataDear Statalisters, I have a panel dataset and I'm struggling to generate the averages of the last 3…
Effect of small treatment sizeDear all, I will please like to find out the impact of having a small treatment size in the analysi…
Subscribe to:
Post Comments (Atom)
0 Response to Swapping male/female as base category in earnings regression model
Post a Comment