Distcomp: Comparing distributions

TL;DR

Distcomp is a statistical command that compares two distributions at each value, controlling false positives, with applications demonstrated through multiple empirical examples.

Contribution

Introduces the distcomp command and methodology for distribution comparison with finite-sample false positive control, illustrated through diverse empirical cases.

Findings

Effective in controlling false positives in distribution comparisons

Applicable to various empirical data analyses

Provides a practical tool for distribution testing

Abstract

The distcomp command is introduced and illustrated. The command assesses whether or not two distributions differ at each possible value while controlling the probability of any false positive, even in finite samples. Syntax and the underlying methodology (from Goldman and Kaplan, 2018) are discussed. Multiple examples illustrate the distcomp command, including revisiting the experimental data of Gneezy and List (2006) and the regression discontinuity design of Cattaneo, Frandsen, and Titiunik (2015).