Estimation and inference of discontinuity in density

Taisuke Otsu, Ke Li Xu, Yukitoshi Matsushita

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Continuity or discontinuity of probability density functions of data often plays a fundamental role in empirical economic analysis. For example, for identification and inference of causal effects in regression discontinuity designs it is typically assumed that the density function of a conditioning variable is continuous at a cutoff point that determines assignment of a treatment. Also, discontinuity in density functions can be a parameter of economic interest, such as in analysis of bunching behaviors of taxpayers. To facilitate researchers to conduct valid inference for these problems, this article extends the binning and local likelihood approaches to estimate discontinuity of density functions and proposes empirical likelihood-based tests and confidence sets for the discontinuity. In contrast to the conventional Wald-type test and confidence set using the binning estimator, our empirical likelihood-based methods (i) circumvent asymptotic variance estimation to construct the test statistics and confidence sets; (ii) are invariant to nonlinear transformations of the parameters of interest; (iii) offer confidence sets whose shapes are automatically determined by data; and (iv) admit higher-order refinements, so-called Bartlett corrections. Firstand second-order asymptotic theories are developed. Simulations demonstrate the superior finite sample behaviors of the proposed methods. In an empirical application, we assess the identifying assumption of no manipulation of class sizes in the regression discontinuity design studied by Angrist and Lavy (1999).

Original languageEnglish
Pages (from-to)507-524
Number of pages18
JournalJournal of Business and Economic Statistics
Volume31
Issue number4
DOIs
Publication statusPublished - 2013 Jan 1
Externally publishedYes

Fingerprint

Confidence Set
Discontinuity
confidence
Density Function
Binning
Empirical Likelihood
Test Set
regression
conditioning
Regression
Bartlett Correction
Local Likelihood
Second-order Asymptotics
manipulation
economics
Economic Analysis
Causal Effect
continuity
Nonlinear Transformation
Variance Estimation

Keywords

  • Bartlett correction
  • Empirical likelihood
  • Local likelihood
  • Nonparametric inference
  • Regression discontinuity design

ASJC Scopus subject areas

  • Statistics and Probability
  • Social Sciences (miscellaneous)
  • Economics and Econometrics
  • Statistics, Probability and Uncertainty

Cite this

Estimation and inference of discontinuity in density. / Otsu, Taisuke; Xu, Ke Li; Matsushita, Yukitoshi.

In: Journal of Business and Economic Statistics, Vol. 31, No. 4, 01.01.2013, p. 507-524.

Research output: Contribution to journalArticle

Otsu, Taisuke ; Xu, Ke Li ; Matsushita, Yukitoshi. / Estimation and inference of discontinuity in density. In: Journal of Business and Economic Statistics. 2013 ; Vol. 31, No. 4. pp. 507-524.
@article{1bf3c383a1cc4ec8a5b912078d97f12b,
title = "Estimation and inference of discontinuity in density",
abstract = "Continuity or discontinuity of probability density functions of data often plays a fundamental role in empirical economic analysis. For example, for identification and inference of causal effects in regression discontinuity designs it is typically assumed that the density function of a conditioning variable is continuous at a cutoff point that determines assignment of a treatment. Also, discontinuity in density functions can be a parameter of economic interest, such as in analysis of bunching behaviors of taxpayers. To facilitate researchers to conduct valid inference for these problems, this article extends the binning and local likelihood approaches to estimate discontinuity of density functions and proposes empirical likelihood-based tests and confidence sets for the discontinuity. In contrast to the conventional Wald-type test and confidence set using the binning estimator, our empirical likelihood-based methods (i) circumvent asymptotic variance estimation to construct the test statistics and confidence sets; (ii) are invariant to nonlinear transformations of the parameters of interest; (iii) offer confidence sets whose shapes are automatically determined by data; and (iv) admit higher-order refinements, so-called Bartlett corrections. Firstand second-order asymptotic theories are developed. Simulations demonstrate the superior finite sample behaviors of the proposed methods. In an empirical application, we assess the identifying assumption of no manipulation of class sizes in the regression discontinuity design studied by Angrist and Lavy (1999).",
keywords = "Bartlett correction, Empirical likelihood, Local likelihood, Nonparametric inference, Regression discontinuity design",
author = "Taisuke Otsu and Xu, {Ke Li} and Yukitoshi Matsushita",
year = "2013",
month = "1",
day = "1",
doi = "10.1080/07350015.2013.818007",
language = "English",
volume = "31",
pages = "507--524",
journal = "Journal of Business and Economic Statistics",
issn = "0735-0015",
publisher = "American Statistical Association",
number = "4",

}

TY - JOUR

T1 - Estimation and inference of discontinuity in density

AU - Otsu, Taisuke

AU - Xu, Ke Li

AU - Matsushita, Yukitoshi

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Continuity or discontinuity of probability density functions of data often plays a fundamental role in empirical economic analysis. For example, for identification and inference of causal effects in regression discontinuity designs it is typically assumed that the density function of a conditioning variable is continuous at a cutoff point that determines assignment of a treatment. Also, discontinuity in density functions can be a parameter of economic interest, such as in analysis of bunching behaviors of taxpayers. To facilitate researchers to conduct valid inference for these problems, this article extends the binning and local likelihood approaches to estimate discontinuity of density functions and proposes empirical likelihood-based tests and confidence sets for the discontinuity. In contrast to the conventional Wald-type test and confidence set using the binning estimator, our empirical likelihood-based methods (i) circumvent asymptotic variance estimation to construct the test statistics and confidence sets; (ii) are invariant to nonlinear transformations of the parameters of interest; (iii) offer confidence sets whose shapes are automatically determined by data; and (iv) admit higher-order refinements, so-called Bartlett corrections. Firstand second-order asymptotic theories are developed. Simulations demonstrate the superior finite sample behaviors of the proposed methods. In an empirical application, we assess the identifying assumption of no manipulation of class sizes in the regression discontinuity design studied by Angrist and Lavy (1999).

AB - Continuity or discontinuity of probability density functions of data often plays a fundamental role in empirical economic analysis. For example, for identification and inference of causal effects in regression discontinuity designs it is typically assumed that the density function of a conditioning variable is continuous at a cutoff point that determines assignment of a treatment. Also, discontinuity in density functions can be a parameter of economic interest, such as in analysis of bunching behaviors of taxpayers. To facilitate researchers to conduct valid inference for these problems, this article extends the binning and local likelihood approaches to estimate discontinuity of density functions and proposes empirical likelihood-based tests and confidence sets for the discontinuity. In contrast to the conventional Wald-type test and confidence set using the binning estimator, our empirical likelihood-based methods (i) circumvent asymptotic variance estimation to construct the test statistics and confidence sets; (ii) are invariant to nonlinear transformations of the parameters of interest; (iii) offer confidence sets whose shapes are automatically determined by data; and (iv) admit higher-order refinements, so-called Bartlett corrections. Firstand second-order asymptotic theories are developed. Simulations demonstrate the superior finite sample behaviors of the proposed methods. In an empirical application, we assess the identifying assumption of no manipulation of class sizes in the regression discontinuity design studied by Angrist and Lavy (1999).

KW - Bartlett correction

KW - Empirical likelihood

KW - Local likelihood

KW - Nonparametric inference

KW - Regression discontinuity design

UR - http://www.scopus.com/inward/record.url?scp=84912029841&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84912029841&partnerID=8YFLogxK

U2 - 10.1080/07350015.2013.818007

DO - 10.1080/07350015.2013.818007

M3 - Article

AN - SCOPUS:84912029841

VL - 31

SP - 507

EP - 524

JO - Journal of Business and Economic Statistics

JF - Journal of Business and Economic Statistics

SN - 0735-0015

IS - 4

ER -