Skip to main content

Table 1 Nonparametric fuzzy regression discontinuity (RD) estimates

From: The effect of retirement on physical and mental health in China: a nonparametric fuzzy regression discontinuity study

 

Without covariates

With covariates

Self-reported health

Life satisfaction

CESD scale

Self-reported health

Life satisfaction

CESD scale

MSE Optimal

 Bandwidth

3.3

3.8

3.6

3.5

3.0

3.2

 Conventional Point Estimate

0.0811

-0.0567

1.4545

0.1018

-0.0475

1.4448

 Conventional 95% CI

[-0.2831, 0.4453]

[-0.3030, 0.1897]

[-3.7613, 6.6703]

[-0.3249, 0.5284]

[-0.3736, 0.2785]

[-5.0521, 7.9418]

 Robust Point Estimate

0.0756

-0.0215

1.2355

0.0913

0.0144

1.0375

 Robust 95% CI

[-0.3579, 0.5090]

[-0.3100, 0.2671]

[-4.8459, 7.3168]

[-0.4193, 0.6018]

[-0.3761, 0.4049]

[-6.6529, 8.7279]

 Effective obs L|R

994|992

1063|1055

1035|1042

1019|1015

867|882

891|907

CER Optimal

 Bandwidth

2.1

2.5

2.4

2.2

2

2.1

 Conventional Point Estimate

0.1856

-0.0208

1.9086

0.2071

-0.0073

1.785

 Conventional 95% CI

[-0.3283, 0.6995]

[-0.3218, 0.2801]

[-4.8901, 8.7072]

[-0.3798, 0.7939]

[-0.3974, 0.3828]

[-6.4640, 10.0339]

 Robust Point Estimate

0.1827

-0.0063

1.7721

0.2025

0.0171

1.5985

 Robust 95% CI

[-0.3664, 0.7319]

[-0.3271, 0.3145]

[-5.4279, 8.9721]

[-0.4266, 0.8316]

[-0.4022, 0.4365]

[-7.1938, 10.3908]

 Effective obs L|R

646|616

724|695

700|665

698|650

567|562

602|586

Obs L|R

3001|2802

2868|2747

2892|2751

3001|2802

2868|2747

2892|2751

  1. (i) All estimates are computed using a triangular Kernel and nearest neighbor heteroskedasticity-robust variance estimator as suggested by CCT framework. (ii) Effective number of observations depends on the size of the data-driven selected bandwidth
  2. All the p-values > 0.05