Weights in csdid

[Zhengjun Xiao]

I am currently working on estimating the treatment effect using a doubly robust DiD method.
My question pertains to the implementation of IPW within this framework. I've calculated weights through a probit model that includes only a part of covariates.
Is it methodologically sound to use these weights as the importance weights in the doubly robust estimation?
I would greatly appreciate any insights or advice on this matter.
Thank you.

[FernandoRios]

You can do that but not sure why you want to
csdid dripw and dript calculate those weights too.
in any case you will also be missing the se estimation that way

[Zhengjun Xiao]

Hello @FernandoRios,
Thank you for your response.

Upon reviewing the yearly standardized differences of covariates using logit regression, I still observed the unbalanced.
This observation led me to consider the probit, which I've found could improve balance. This is the reason I asked the question.

If I can replace the weights, given the staggered timing in my data, I'm considering whether it would be appropriate to apply the weights only prior to the earliest treatment year (for example, the year 2000), without considering the other treated years such as 2001 or 2002.

Thanks,
Xiao