Skip to main content

Table 2 Regression Results for Hospital Bed Density, China Counties, 2011

From: Geographic distribution of hospital beds throughout China: a county-level econometric analysis

 

Log of hospital beds per 1000 people

OLS models

Spatial lag models

Spatial error models

Variables

(1)

(2)

(3)

(4)

(5)

(6)

Log of saving per capita

0.175***

0.228***

0.144***

0.216***

0.233***

0.232***

(10.693)

(12.985)

(8.852)

(12.479)

(13.321)

(13.321)

Log of government revenue per capita

0.107***

0.119***

0.096***

0.110***

0.107***

0.105***

(8.593)

(9.216)

(8.083)

(8.647)

(8.622)

(8.254)

Percentage of urban population (%)

−0.003***

−0.003***

−0.003***

−0.002***

−0.003***

−0.002***

(−4.180)

(−3.073)

(−3.752)

(−2.905)

(−4.080)

(−2.154)

Log of area size (km2)

0.082***

0.034***

0.030***

0.013

−0.009

−0.006

(8.332)

(2.981)

(3.167)

(1.163)

(−0.788)

(−0.477)

Rho (ρ) / Lambda (λ)

  

0.724***

0.486***

0.905***

0.925***

  

(24.329)

(8.751)

(33.334)

(40.683)

Provinces dummies

No

Yes

No

Yes

No

Yes

Sample size

2043

2043

2043

2043

2043

2043

R2

0.253

0.394

0.329

0.411

0.403

0.439

Log likelihood

−936.862

−723.352

−839.045

−698.927

−731.663

−669.284

Akaike info criterion

1883.72

1516.7

1690.09

1469.85

1473.33

1408.57

Spatial model diagnostic tests

Moran’s I (error)

43.144***

19.880***

    

Robust Lagrange Multiplier (lag)

43.214***

0.424

    

Robust Lagrange Multiplier (error)

1378.343***

115.737***

    
  1. (1) ***,** and * denote 1, 5 and 10 % significance levels, respectively; (2) Since there are 31 provinces in our dataset, 30 provinces dummies are added to the corresponding regressions; (3) All spatial model diagnostic tests are based on OLS