Local Polynomial Order in Regression Discontinuity Designs

被引:27
作者
Pei, Zhuan [1 ,2 ]
Lee, David S. [3 ,4 ]
Card, David [2 ,4 ,5 ]
Weber, Andrea [2 ,6 ]
机构
[1] Cornell Univ, Dept Policy Anal & Management, Ithaca, NY 14853 USA
[2] IZA Inst Lab Econ, Bonn, Germany
[3] Princeton Univ, Dept Econ, Princeton, NJ 08544 USA
[4] Natl Bur Econ Res, Cambridge, MA 02138 USA
[5] Univ Calif Berkeley, Dept Econ, Berkeley, CA USA
[6] Cent European Univ, Dept Econ & Business, Budapest, Hungary
关键词
Local polynomial estimation; Polynomial order; Regression discontinuity design; Regression kink design; OPTIMAL BANDWIDTH CHOICE; HEAD-START; INFERENCE; SELECTION;
D O I
10.1080/07350015.2021.1920961
中图分类号
F [经济];
学科分类号
02 ;
摘要
Treatment effect estimates in regression discontinuity (RD) designs are often sensitive to the choice of bandwidth and polynomial order, the two important ingredients of widely used local regression methods. While Imbens and Kalyanaraman and Calonico, Cattaneo, and Titiunik provided guidance on bandwidth, the sensitivity to polynomial order still poses a conundrum to RD practitioners. It is understood in the econometric literature that applying the argument of bias reduction does not help resolve this conundrum, since it would always lead to preferring higher orders. We therefore extend the frameworks of Imbens and Kalyanaraman and Calonico, Cattaneo, and Titiunik and use the asymptotic mean squared error of the local regression RD estimator as the criterion to guide polynomial order selection. We show in Monte Carlo simulations that the proposed order selection procedure performs well, particularly in large sample sizes typically found in empirical RD applications. This procedure extends easily to fuzzy regression discontinuity and regression kink designs.
引用
收藏
页码:1259 / 1267
页数:9
相关论文
共 50 条
[41]   Simultaneous selection of optimal bandwidths for the sharp regression discontinuity estimator [J].
Arai, Yoichi ;
Ichimura, Hidehiko .
QUANTITATIVE ECONOMICS, 2018, 9 (01) :441-482
[42]   Optimal bandwidth choice for robust bias-corrected inference in regression discontinuity designs [J].
Calonico, Sebastian ;
Cattaneo, Matias D. ;
Farrell, Max H. .
ECONOMETRICS JOURNAL, 2020, 23 (02) :192-210
[43]   Approaches to the Estimation of the Local Average Treatment Effect in a Regression Discontinuity Design [J].
O'Keeffe, Aidan G. ;
Baio, Gianluca .
SCANDINAVIAN JOURNAL OF STATISTICS, 2016, 43 (04) :978-995
[44]   Analysis of regression-discontinuity designs with multiple cutoffs or multiple scores [J].
Cattaneo, Matias D. ;
Titiunik, Rocio ;
Vazquez-Bare, Gonzalo .
STATA JOURNAL, 2020, 20 (04) :866-891
[45]   Nonlinear and nonseparable structural functions in regression discontinuity designs with a continuous treatment [J].
Xie, Haitian .
JOURNAL OF ECONOMETRICS, 2024, 242 (01)
[46]   Extrapolating Treatment Effects in Multi-Cutoff Regression Discontinuity Designs [J].
Cattaneo, Matias D. ;
Keele, Luke ;
Titiunik, Rocio ;
Vazquez-Bare, Gonzalo .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2021, 116 (536) :1941-1952
[47]   Are authenticity labels effective for local tourism efficiency? A regression discontinuity design approach [J].
Di Matteo, Dante ;
Ferrara, Antonella Rita .
JOURNAL OF CULTURAL ECONOMICS, 2025,
[48]   Binary quantile regression with local polynomial smoothing [J].
Chen, Songnian ;
Zhang, Hanghui .
JOURNAL OF ECONOMETRICS, 2015, 189 (01) :24-40
[49]   LOCAL POLYNOMIAL AND PENALIZED TRIGONOMETRIC SERIES REGRESSION [J].
Huang, Li-Shan ;
Chan, Kung-Sik .
STATISTICA SINICA, 2014, 24 (03) :1215-1238
[50]   A Correction for Regression Discontinuity Designs With Group-Specific Mismeasurement of the Running Variable [J].
Bartalotti, Otavio ;
Brummet, Quentin ;
Dieterle, Steven .
JOURNAL OF BUSINESS & ECONOMIC STATISTICS, 2021, 39 (03) :833-848