TY - JOUR
T1 - Local Polynomial Order in Regression Discontinuity Designs
AU - Pei, Zhuan
AU - Lee, David S.
AU - Card, David
AU - Weber, Andrea
N1 - Publisher Copyright:
© 2021 The Author(s). Published with license by Taylor & Francis Group, LLC.
PY - 2022
Y1 - 2022
N2 - 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.
AB - 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.
KW - Local polynomial estimation
KW - Polynomial order
KW - Regression discontinuity design
KW - Regression kink design
UR - http://www.scopus.com/inward/record.url?scp=85107414136&partnerID=8YFLogxK
U2 - 10.1080/07350015.2021.1920961
DO - 10.1080/07350015.2021.1920961
M3 - Article
AN - SCOPUS:85107414136
SN - 0735-0015
VL - 40
SP - 1259
EP - 1267
JO - Journal of Business and Economic Statistics
JF - Journal of Business and Economic Statistics
IS - 3
ER -