Regression Discontinuity with Integer Score and Non-Integer Cutoff

@article{ART002915967},
author={ 이명재 and 심혜정 and 박상수 },
title={Regression Discontinuity with Integer Score and Non-Integer Cutoff},
journal={The Korean Economic Review},
issn={0254-3737},
year={2023},
volume={39},
number={1},
pages={73-101},
doi={10.22841/kerdoi.2023.39.1.003},
url={http://dx.doi.org/10.22841/kerdoi.2023.39.1.003}

TY - JOUR
AU - 이명재
AU - 심혜정
AU - 박상수
TI - Regression Discontinuity with Integer Score and Non-Integer Cutoff
T2 - The Korean Economic Review
PY - 2023
VL - 39
IS - 1
PB - 한국경제학회
SP - 73-101
SN - 0254-3737
AB - In regression discontinuity (RD), the treatment is determined by a continuous score G crossing a cutoff c or not. However, often G is observed only as the ‘rounded-down integer S’ (e.g., birth year observed instead of birth time), and c is not an integer. In this case, the “cutoff sample” (i.e., the observations with S equal to the rounded-down integer of c) is discarded due to the ambiguity in G crossing c or not. We show that, first, if the usual RD estimators are used with the integer nature of S ignored, then a bias occurs, but it becomes zero if a slope symmetry condition holds or if c takes a certain “middle” value. Second, the distribution of the measurement error e G S = - can be specified and tested for, and if the distribution is accepted, then the cutoff sample can be used fruitfully. Third, two-step estimators and bootstrap inference are available in the literature, but a single-step ordinary least squares or instrumental variable estimator is enough. We also provide a simulation study and an empirical analysis for a dental support program based on age in South Korea.
KW - Regression Discontinuity, Integer Running Variable, Non-integer Cutoff
DO - 10.22841/kerdoi.2023.39.1.003
UR - http://dx.doi.org/10.22841/kerdoi.2023.39.1.003
ER -

이명재 , 심혜정 and 박상수 (2023). Regression Discontinuity with Integer Score and Non-Integer Cutoff. The Korean Economic Review, 39( 1), 73- 101.

이명재 , 심혜정 and 박상수 . 2023, “Regression Discontinuity with Integer Score and Non-Integer Cutoff”, The Korean Economic Review, vol. 39, no. 1, pp. 73-101. Available from: doi:10.22841/kerdoi.2023.39.1.003

이명재 심혜정 et al. 박상수 “Regression Discontinuity with Integer Score and Non-Integer Cutoff” The Korean Economic Review 39.1 pp. 73-101 (2023): 73.

이명재 , 심혜정 , 박상수 . Regression Discontinuity with Integer Score and Non-Integer Cutoff The Korean Economic Review [Internet]. 2023; 39( 1), : 73-101. Available from: doi:10.22841/kerdoi.2023.39.1.003

이명재 , 심혜정 and 박상수 . “Regression Discontinuity with Integer Score and Non-Integer Cutoff” The Korean Economic Review 39, no.1 (2023): 73-101. doi: 10.22841/kerdoi.2023.39.1.003

홈 권호 목록 논문 목록 논문 상세

Regression Discontinuity with Integer Score and Non-Integer Cutoff

The Korean Economic Review

약어 : KER

2023, vol.39, no.1, pp.73 - 101

DOI : 10.22841/kerdoi.2023.39.1.003

발행기관 : 한국경제학회

연구분야 : 경제학

이명재¹ , 심혜정² , 박상수³

¹고려대학교

²한국무역협회

³고려대학교

인용한 논문 수 : - 서지 간략 보기

초록

In regression discontinuity (RD), the treatment is determined by a continuous score G crossing a cutoff c or not. However, often G is observed only as the ‘rounded-down integer S’ (e.g., birth year observed instead of birth time), and c is not an integer. In this case, the “cutoff sample” (i.e., the observations with S equal to the rounded-down integer of c) is discarded due to the ambiguity in G crossing c or not. We show that, first, if the usual RD estimators are used with the integer nature of S ignored, then a bias occurs, but it becomes zero if a slope symmetry condition holds or if c takes a certain “middle” value. Second, the distribution of the measurement error e G S = - can be specified and tested for, and if the distribution is accepted, then the cutoff sample can be used fruitfully. Third, two-step estimators and bootstrap inference are available in the literature, but a single-step ordinary least squares or instrumental variable estimator is enough. We also provide a simulation study and an empirical analysis for a dental support program based on age in South Korea.