Diff command - Kernel-based Propensity Score Matching DID in the context of RCS data

[Jaya Jumrani]

Hello,

Hope you all are keeping safe. I hope some of you would be better positioned to answer a few of my queries in the context of diff command (developed by Juan M. Villa and available on SSC) that can perform PSM and DID analyses.

1. How are the standard errors computed in the context of kernel PSM for repeated cross-sectional (rcs) data? The help file states that under certain circumstances parametric estimation can be problematic and thus bootstrapped standard errors can be employed as an alternative. What are such circumstances? It would be nice if any of you could elaborate a bit on that and suggest an apt reference(s) for the same.
2. Can the number of observations vary with the use of sampling weights in the context of rcs? If so, why?

Thanks in anticipation.

Best,
Jaya