[Anjali Kajan]

Yes I thought it had to be LPM because my dependent variable is binary (whether married under 18 or not) but the dchildmar#postreform2 interaction is my triple difference estimator

[Carlo Lazzaro]

Anjali:
see if -margins- can be of some help.

[Dilini Jayasinghe]

Hi,

I'm new to Statalist and I hope it's alright to post a follow-up question. I am using xtdidregress command to identify the causal effect of a policy on school achievement. My control group is students in Victoria (Australia) and treated group is students in New South Wales. I want to know if I should include time invariant controls in this did regression model as they would anyway get wiped out by the did regression

My command is as follows:

xtset hicid year
xtdidregress (schl_achieve post exposed x1) (did), group(state) time(year)

where x1 = time varying controls

Thanks
Dilini

[Clyde Schechter]

-xtdidregress- performs a fixed-effects regression. If you include variables that are attributes of the groups that do not vary over time, they will be colinear with the group fixed effects an will be dropped from the analysis. So there is no point in including them.

And don't forget that one of the strengths of the fixed-effects regression is that the effects of any such time-invariant attributes of the groups are automatically accounted for by the fixed-effects regression, even if they are not included in the regression model (and even if there is no data available about them, and even if you don't know these variables exist). So the non-inclusion or omission of these variables has no consequence.

[Dilini Jayasinghe]

Thank you very much !