5.2: Results Of Panel Data Analysis
5.2. Results of Panel Data Analysis
Table 3 presents our estimation results of the effects of well-being facilities on elderly subjective well-being based on the equation described in the Section 3.1. In model 1, we only include our key variable, the number of well-being facilities, using the OLS model. We include other control variables in model 2. In order to control for the time and location specific characteristics, we add the year fixed-effects in model 3, and year and region fixed-effects in model 4, respectively. The estimation results are constant in model 1 through model 3. That is, the effects of the number of well-fare facilities positively influence elderly subjective well-being. But, when we add the region-specific fixed-effects as shown in model 4, the variable of the number of well-fare facilities is not statistically significant. As discussed above, this inconsistent result may be due to the regional unobserved characteristics. In order to account for it, we use two-stage least square (2SLS) estimator with an instrumental variable, such as the number of elderly population in each region. Model 5 presents the estimation result of 2SLS, suggesting that the number of well-fare facilities in a local region has a positive effect on elderly subjective well-being. The result demonstrates that a provision of the well-fare facilities at the local level would improve happiness of older adults. …show more content…
Empirical Results of Panel Data Analysis OLS 2SLS (1) (2) (3) (4) (5)
Well-fare Facilities 0.0026 *** 0.0016 *** 0.0004 *** 0.0001 0.0137 *** (0.0001 ) (0.0001 ) (0.0001 ) (0.0010 ) (0.0020 )
Married 0.5708 *** 0.2276 *** 0.2238 *** 0.2227 *** (0.0130 ) (0.0169 ) (0.0169 ) (0.0163
Table 3 presents our estimation results of the effects of well-being facilities on elderly subjective well-being based on the equation described in the Section 3.1. In model 1, we only include our key variable, the number of well-being facilities, using the OLS model. We include other control variables in model 2. In order to control for the time and location specific characteristics, we add the year fixed-effects in model 3, and year and region fixed-effects in model 4, respectively. The estimation results are constant in model 1 through model 3. That is, the effects of the number of well-fare facilities positively influence elderly subjective well-being. But, when we add the region-specific fixed-effects as shown in model 4, the variable of the number of well-fare facilities is not statistically significant. As discussed above, this inconsistent result may be due to the regional unobserved characteristics. In order to account for it, we use two-stage least square (2SLS) estimator with an instrumental variable, such as the number of elderly population in each region. Model 5 presents the estimation result of 2SLS, suggesting that the number of well-fare facilities in a local region has a positive effect on elderly subjective well-being. The result demonstrates that a provision of the well-fare facilities at the local level would improve happiness of older adults. …show more content…
Empirical Results of Panel Data Analysis OLS 2SLS (1) (2) (3) (4) (5)
Well-fare Facilities 0.0026 *** 0.0016 *** 0.0004 *** 0.0001 0.0137 *** (0.0001 ) (0.0001 ) (0.0001 ) (0.0010 ) (0.0020 )
Married 0.5708 *** 0.2276 *** 0.2238 *** 0.2227 *** (0.0130 ) (0.0169 ) (0.0169 ) (0.0163