Minimum distance estimator for sharp regression discontinuity with multiple running variables

Jin young Choi, Myoung-jae Lee

Research output: Contribution to journalArticle

Abstract

In typical regression discontinuity, a running variable (or ‘score’) crosses a cutoff to determine a treatment. There are, however, many regression discontinuity cases where multiple scores have to cross all of their cutoffs to get treated. One approach to deal with these cases is one-dimensional localization using a single score on the subpopulation with all the other scores already crossing the cutoffs (“conditional one-dimensional localization approach, CON”), which is, however, inconsistent when partial effects are present which occur when some, but not all, scores cross their cutoffs. Another approach is multi-dimensional localization explicitly allowing for partial effects, which is, however, less efficient than CON due to more localizations than in CON. We propose a minimum distance estimator that is at least as efficient as CON, yet consistent even when partial effects are present. A simulation study demonstrates these characteristics of the minimum distance estimator.

Original languageEnglish
Pages (from-to)10-14
Number of pages5
JournalEconomics Letters
Volume162
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

Estimator
Regression discontinuity
Minimum distance
Localization
Simulation study

Keywords

  • Minimum distance estimator
  • Multiple running variables
  • Regression discontinuity

ASJC Scopus subject areas

  • Finance
  • Economics and Econometrics

Cite this

Minimum distance estimator for sharp regression discontinuity with multiple running variables. / Choi, Jin young; Lee, Myoung-jae.

In: Economics Letters, Vol. 162, 01.01.2018, p. 10-14.

Research output: Contribution to journalArticle

@article{1ff23f3341a7449592e2dee124fe9438,
title = "Minimum distance estimator for sharp regression discontinuity with multiple running variables",
abstract = "In typical regression discontinuity, a running variable (or ‘score’) crosses a cutoff to determine a treatment. There are, however, many regression discontinuity cases where multiple scores have to cross all of their cutoffs to get treated. One approach to deal with these cases is one-dimensional localization using a single score on the subpopulation with all the other scores already crossing the cutoffs (“conditional one-dimensional localization approach, CON”), which is, however, inconsistent when partial effects are present which occur when some, but not all, scores cross their cutoffs. Another approach is multi-dimensional localization explicitly allowing for partial effects, which is, however, less efficient than CON due to more localizations than in CON. We propose a minimum distance estimator that is at least as efficient as CON, yet consistent even when partial effects are present. A simulation study demonstrates these characteristics of the minimum distance estimator.",
keywords = "Minimum distance estimator, Multiple running variables, Regression discontinuity",
author = "Choi, {Jin young} and Myoung-jae Lee",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.econlet.2017.10.002",
language = "English",
volume = "162",
pages = "10--14",
journal = "Economics Letters",
issn = "0165-1765",
publisher = "Elsevier",

}

TY - JOUR

T1 - Minimum distance estimator for sharp regression discontinuity with multiple running variables

AU - Choi, Jin young

AU - Lee, Myoung-jae

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In typical regression discontinuity, a running variable (or ‘score’) crosses a cutoff to determine a treatment. There are, however, many regression discontinuity cases where multiple scores have to cross all of their cutoffs to get treated. One approach to deal with these cases is one-dimensional localization using a single score on the subpopulation with all the other scores already crossing the cutoffs (“conditional one-dimensional localization approach, CON”), which is, however, inconsistent when partial effects are present which occur when some, but not all, scores cross their cutoffs. Another approach is multi-dimensional localization explicitly allowing for partial effects, which is, however, less efficient than CON due to more localizations than in CON. We propose a minimum distance estimator that is at least as efficient as CON, yet consistent even when partial effects are present. A simulation study demonstrates these characteristics of the minimum distance estimator.

AB - In typical regression discontinuity, a running variable (or ‘score’) crosses a cutoff to determine a treatment. There are, however, many regression discontinuity cases where multiple scores have to cross all of their cutoffs to get treated. One approach to deal with these cases is one-dimensional localization using a single score on the subpopulation with all the other scores already crossing the cutoffs (“conditional one-dimensional localization approach, CON”), which is, however, inconsistent when partial effects are present which occur when some, but not all, scores cross their cutoffs. Another approach is multi-dimensional localization explicitly allowing for partial effects, which is, however, less efficient than CON due to more localizations than in CON. We propose a minimum distance estimator that is at least as efficient as CON, yet consistent even when partial effects are present. A simulation study demonstrates these characteristics of the minimum distance estimator.

KW - Minimum distance estimator

KW - Multiple running variables

KW - Regression discontinuity

UR - http://www.scopus.com/inward/record.url?scp=85033576959&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85033576959&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2017.10.002

DO - 10.1016/j.econlet.2017.10.002

M3 - Article

VL - 162

SP - 10

EP - 14

JO - Economics Letters

JF - Economics Letters

SN - 0165-1765

ER -