Skip to main content

Table 7 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

Smoking

Drinking

Physical Activity

Smoking

Drinking

Physical Activity

MSE Optimal

 Bandwidth

3.5

3.6

3.1

3.6

3.5

3.5

 Point Estimate

-0.0434

-0.0901

-0.0596

0.2871

0.1798

-0.0702

 Conventional 95% CI

[-0.4578, 0.3710]

[-0.5639, 0.3836]

[-0.2505, 0.1313]

[-0.1567, 0.7309]

[-0.3496, 0.7092]

[-0.2849, 0.1444]

 Robust Point Estimate

-0.0597

-0.0953

-0.0563

0.3298

0.2192

-0.0902

 Robust 95% CI

[-0.5481, 0.4287]

[-0.6566, 0.4660]

[-0.2773, 0.1647]

[-0.1914, 0.8509]

[-0.4096, 0.8479]

[-0.3367, 0.1564]

 Effective obs L|R

1032|1001

1140|1109

583|596

1032|1001

1087|1057

667|689

CER Optimal

 Bandwidth

2.3

2.3

2

2.3

2.2

2.3

 Point Estimate

-0.0883

-0.1436

-0.0370

0.1776

0.0844

-0.0536

 Conventional 95% CI

[-0.6237, 0.4471]

[-0.7612, 0.4740]

[-0.2547, 0.1807]

[-0.3730, 0.7283]

[-0.5904, 0.7593]

[-0.2884, 0.1812]

 Robust Point Estimate

-0.0931

-0.1435

-0.0370

0.1965

0.1012

-0.0634

 Robust 95% CI

[-0.6636, 0.4774]

[-0.8022, 0.5152]

[-0.2691, 0.1950]

[-0.3886, 0.7817]

[-0.6199, 0.8223]

[-0.3127, 0.1860]

 Effective obs L|R

721|653

769|708

406|402

721|653

743|678

476|446

Obs L|R

2998|2700

3191|2924

1806|1866

2998|2700

3191|2924

1806|1866

  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 depend on the size of the data-driven selected bandwidth
  2. All the p-values > 0.05