{"id":"arxiv/2610.31978","url":"https://arxiv.org/abs/2610.31978","tags":["stat.ME","econ.EM","math.ST"],"score":0,"title":"Parallel Trends as a Spectrum","domain":"stat.ME","source":"arxiv","authors":["Kyle Schindl","Joseph Fry","Florian Gunsilius","Larry Wasserman"],"reviews":0,"subject":"stat.ME","summary":"Difference-in-differences (DiD) designs are typically used to identify and estimate average causal effects, even though treatment can affect the entire outcome distribution. This has led to a growing interest in distributional DiD methods. However, identifying the full counterfactual distribution requires a parallel trends assumption on the entire distributional dynamics, which can be too strong in practice. We address this problem by introducing a spectrum of identifying restrictions that lie between the average and fully distributional DiD by imposing parallel trends on $L$-moments. This…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-10-22","reactions":{}}