Sitakanta Panda
@stkntp.bsky.social
Interests in political & public economics, labour markets, law and economics, development. Cinema/literature/social science buff. AstProf, IIT Bhubaneswar. https://sites.google.com/site/spandaecon
Reposted by Sitakanta Panda
We show:
* LP-DiD is identical to "stacked DiD" (Cengiz et al. 2019); but no need to stack the data
* a simple re-weighted LP-DiD estimate is identical to Callaway/Sant'Anna; but uses OLS & asymptotic inference: much faster.
* slightly modified LP-DiD closely related to of Borusyak et al.
* LP-DiD is identical to "stacked DiD" (Cengiz et al. 2019); but no need to stack the data
* a simple re-weighted LP-DiD estimate is identical to Callaway/Sant'Anna; but uses OLS & asymptotic inference: much faster.
* slightly modified LP-DiD closely related to of Borusyak et al.
July 20, 2025 at 3:12 AM
We show:
* LP-DiD is identical to "stacked DiD" (Cengiz et al. 2019); but no need to stack the data
* a simple re-weighted LP-DiD estimate is identical to Callaway/Sant'Anna; but uses OLS & asymptotic inference: much faster.
* slightly modified LP-DiD closely related to of Borusyak et al.
* LP-DiD is identical to "stacked DiD" (Cengiz et al. 2019); but no need to stack the data
* a simple re-weighted LP-DiD estimate is identical to Callaway/Sant'Anna; but uses OLS & asymptotic inference: much faster.
* slightly modified LP-DiD closely related to of Borusyak et al.