Estimating the distribution of treatment effects in difference-in-differences
This paper proposes a new approach to identify the distributional treatment effect parameters in differences-in-differences based on a combination of a random coefficients model with deconvolution.