Researchers release v2 of inference paper on quantile treatment effects with small treatment groups

Marina Dias and Demian Pouzo released a revised version of their paper on inference for conditional quantile treatment effects with small treatment groups, proposing an inference method for conditional quantile treatment effects adapted to small treatment group scenarios. This method combines estimation of the conditional quantile function from a large reference group with an exact conditional Bernoulli representation of quantile indicators, achieving asymptotic size control and deriving power bounds under the premise that the overall sample expands while the treatment group size remains bounded. It can accommodate non-separable outcome functions, heteroscedasticity, and flexible error distributions across different treatment configurations, and is applicable to dynamic treatment models, non-separable quantile models, and staggered difference-in-differences adoption designs.

AI-generated from reports · 1 hr agoUpdated

Not enough consecutive observations yet; trend is not drawn.