Testing Density Discontinuity Using Spacings

In the regression discontinuity design, researchers assess the credibility of assumptions by testing the running variable's density for a discontinuity at the cutoff. This paper proposes a novel such test. The test uses q spacings on each side of the cutoff, where spacings are the widths between ordered observations. Spacings yield a continuous statistic that behaves well under weak assumptions. I study its properties in two asymptotic frameworks: one where q is fixed as the sample size increases and one where q slowly diverges to infinity. With q fixed, the test achieves exact asymptotic size control without user randomization. When q increases slowly, the test becomes consistent without sacrificing exact asymptotic size. Compared to the preferred version of Bugni and Canay's (2021) test, the only other test with comparably weak assumptions, this test reduces sensitivity to the tuning parameter and performs similarly when theirs is well-chosen.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-12
DOI
https://doi.org/10.48550/arxiv.2611.16199
Primary Topic
Advanced Causal Inference Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Testing Density Discontinuity Using Spacings

William T. Firmin
arXiv (Cornell University)
Advanced Causal Inference Techniques
preprint

Testing Density Discontinuity Using Spacings

William T. Firmin
preprint en

Abstract

In the regression discontinuity design, researchers assess the credibility of assumptions by testing the running variable's density for a discontinuity at the cutoff. This paper proposes a novel such test. The test uses q spacings on each side of the cutoff, where spacings are the widths between ordered observations. Spacings yield a continuous statistic that behaves well under weak assumptions. I study its properties in two asymptotic frameworks: one where q is fixed as the sample size increases and one where q slowly diverges to infinity. With q fixed, the test achieves exact asymptotic size control without user randomization. When q increases slowly, the test becomes consistent without sacrificing exact asymptotic size. Compared to the preferred version of Bugni and Canay's (2021) test, the only other test with comparably weak assumptions, this test reduces sensitivity to the tuning parameter and performs similarly when theirs is well-chosen.

arXiv (Cornell University)
Advanced Causal Inference Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon