In this research, we could obtain several results which fit for the predictions comes from our model setting. However, there remain some issues.
The first point is to introduce more explaining variables and samples to improve the fitting of the model since we used only 115 panel samples.
The second point is to use individual data instead of the macro data used here. Since we used macro data, we can’t make much analysis on the differences between male and female, between age, and individual life styles. In order to know the effects of those factors much, we should use the individual data.
The third point is to confirm the cross effect of aging and other factors. For the simplifica-tion, variables of income, educasimplifica-tion, and so on, are set to be exogenous variables in this paper.
By using simultaneous equations and simultaneous estimation method, we should deepen our analysis.
References
[1] Agnes l. F. Chan, Teng Chieh Yang, Jian-xun Chen, Lee Haw Yu, Henry W. C. Leung (2006) ”Cost of Depression of Adults in Taiwan,” The International Journal of Psychiatry in Medicine. Vol.36, No.1, pp.131-135.
[2] Department of Health in Taiwan (2005) National Health Interview Survey in Taiwan.
[3] Department of Health in Taiwan (2009) Statistics on Causes of Deaths in Taiwan.
[4] Department of Health in Taiwan (2005-2009) National Health Expenditures in Taiwan.
[5] Antonio Rodrguez Andre’s (2005) ”Income inequality, unemployment, and suicide: a panel data analysis of 15 European countries,” Applied Economics, 2005, 37, 439-451.
[6] Holmes T.H., Rahe R.H. (1967). ”The Social Readjustment Rating Scale,” Journal of Psy-chosomatic Research, Vol.11, No.2, pp.213-218.
Table 1: Prevalence of mental disorders age 12-64
Aware symptoms Inform health carers total number ratio total ratio
total 18711 488 2.6 330 68.6
Sex
male 9470 217 2.3 147 67.8
female 9241 271 2.9 183 67.5
Age
12-19 2840 38 1.3 20 52.2
20-34 6192 125 2 75 60.4
35-44 4149 137 3.3 99 72.4
45-64 5530 187 3.4 135 72
Sex by age
Male 12-19 1461 17 1.2 8 46.8
20-34 3156 49 1.6 33 67.5
35-44 2099 61 2.9 46 74.9
45-64 2754 89 3.2 60 67.1
Female 12-19 1380 21 1.5 12 56.7
20-34 3036 75 2.5 42 55.7
35-44 2050 76 3.7 54 70.4
45-64 2776 98 3.5 75 76.5
Urbanization
high 4403 161 3.7 103 64.3
mid 10508 227 2.2 155 68.1
low 3800 99 2.6 71 71.7
Sex by Urban
male high 2157 66 3.1 46 69.4
mid 5287 94 1.8 65 69.2
low 2026 56 2.8 36 63.5
Female high 2247 94 4.2 57 60.7
mid 5221 133 2.6 90 67.4
low 1773 43 2.4 35 82.5
Income
-30,000 3330 142 4.3 101 70.8
30,000-50,000 4152 87 2.1 63 73.1
50,000-70.000 3993 90 2.3 51 56.4
70,000-100,000 3170 71 2.2 48 67.8
100,000- 3141 76 2.4 53 70.5
Unknown 926 21 2.3 13 60.4
Source: Table A-14, National Health Interview Survey in Taiwan (2005).
Table 2: Prevalence of mental disorders age 65+
Aware symptoms Inform health carers total number ratio total ratio
total 2325 96 4.1 64 67
Sex
male 1149 35 3.1 22 66.2
female 1176 61 5.2 41 67.5
Urbanization
high 492 18 3.6 12 68.9
mid 1097 60 5.5 39 66.1
low 736 19 2.5 13 68.3
Sex by Urban
male high 253 2 0.9 1 47.6
mid 533 26 4.9 18 74.1
low 363 7 1.9 3 44
Female high 239 15 6.4 11 72.1
mid 564 34 6 20 60.2
low 373 12 3.2 10 82.4
Income
-30,000 1175 45 3.8 23 51.9
30,000-50,000 433 22 5 17 76.2
50,000-70.000 235 8 3.3 5 62.8
70,000-100,000 133 6 4.4 6 100
100,000- 123 6 5.1 5 72.2
Unknown 225 10 4.3 9 100
Source: Table C-23 in National Health Interview Survey in Taiwan (2005).
Table 3: Death from suicide statistics
Area Number Rate Area Number Rate
Total 4063 17.6052 Taichung City 243 15.5766
Taipei City 341 13.0393 Changhua County 213 16.2261
Kaohsiung City 287 18.7978 Nantou County 113 21.2690
Keelung City 109 28.0458 Yunlin County 123 17.0069
Hsinchu City 70 17.1367 Chiayi County 103 18.7880
Taichung City 169 15.7961 Tainan County 245 22.1830
Tainan City 142 18.4474 Kaohsiung County 226 18.1790
Chiayi City 49 17.8945 Pingtung County 165 18.6707
Taipei County 667 17.3081 Penghu County 20 21.1062
Taoyuan County 337 17.1176 Hualien County 70 20.5159 Hsinchu County 85 16.7627 Taichung County 51 21.9664
Yilan County 99 21.4628 Kinmen County 9 10.0912
Miaoli County 126 22.4571 Lienchiang County 1 10.1657 Source: Table 59 in Statistics on Causes of Deaths in Taiwan (2009), the Department of Health.
Table 4: Medical expenditure for neurotic disorders
Medical cost points Population Per-capita cost
Thousand points Thousand points
Source: Code 05010501 in National Health Expenditures (2009), the Department of Health in Taiwan.
Table 5: Expecting effects on the mental disorder sand expenditure for its medical treatment
Items Effects
Table 6: Basic regression result
Dependent variable: PER_CAPITA_MEDICAL_EXPENSES Current sample: 1 to 115
Number of observations: 115
Mean of dep. var. = 120.1 LM het. test = .3944 [.530]
Std. dev. of dep. var. = 22.78
Sum of squared residuals = .3887E+05 Jarque-Bera test = 45.18 [.000]
Variance of residuals = 363.3 Ramsey’s RESET2 = 2.879 [.093]
Std. error of regression = 19.06 F (zero slopes) = 7.974 [.000]
R-squared = .3428 Schwarz B.I.C. = 517.0 Adjusted R-squared = .2998 Log likelihood = -498.0
---Estimated Standard
Variable Coefficient Error t-statistic P-value
---AGING_RATE 5.655 1.258 4.494 *** [.000]
PER_HOUSE_INCOME -.9396E-04 .5738E-04 -1.637 [.104]
UNEMPLOYMENT_RATE 8.863 3.244 2.732 *** [.007]
MEDICAL_AVAILABILITY 101.0 18.06 5.592 *** [.000]
HIGH_EDU_RATE -4.488 3.114 -1.441 [.152]
DIVORCE_RATE -1.312 5.700 -.2302 [.818]
POPULATION_SIZE .7288E-05 .2802E-05 2.601 ** [.011]
C 19.40 31.40 .6180 [.538]
---Method of estimation = Ordinary Least Squares
Table 7: Regression with logged terms
Dependent variable: LNPER_CAPITA_MEDICAL_EXPENSES Current sample: 1 to 115
Number of observations: 115
Mean of dep. var. = 4.771 LM het. test = 5.143 [.023]
Std. dev. of dep. var. = .1880
Sum of squared residuals = 2.657 Jarque-Bera test = 4.617 [.099]
Variance of residuals = .0248 Ramsey’s RESET2 = 1.068 [.304]
Std. error of regression = .1576 F (zero slopes) = 7.896 [.000]
R-squared = .3406 Schwarz B.I.C. = -34.48 Adjusted R-squared = .2975 Log likelihood = 53.46
---Estimated Standard
Variable Coefficient Error t-statistic P-value
---LNAGING_RATE .2244 .1197 1.876 * [.063]
LNPER_HOUSE_INCOME -.0500 .1293 -.3865 [.700]
LNUNEMPLOYMENT_RATE .2528 .1272 1.988 ** [.049]
LNMEDICAL_AVAILABILITY .3324 .0642 5.181 *** [.000]
LNHIGH_EDU_RATE .1783E-02 .1169 .0153 [.988]
LNDIVORCE_RATE -.1890 .1253 -1.509 [.134]
LNPOPULATION_SIZE -.0179 .0243 -.7375 [.462]
C 5.236 1.623 3.227 ** [.002]
---Method of estimation = Ordinary Least Squares
Table 8: Regression with square terms
Dependent variable: PER_CAPITA_MEDICAL_EXPENSES Current sample: 1 to 115
Number of observations: 115
Mean of dep. var. = 120.1 LM het. test = 6.890 [.009]
Std. dev. of dep. var. = 22.78
Sum of squared residuals = .2869E+05 Jarque-Bera test = 14.88 [.001]
Variance of residuals = 275.8 Ramsey’s RESET2 = .3761 [.541]
Std. error of regression = 16.61 F (zero slopes) = 11.04 [.000]
R-squared = .5150 Schwarz B.I.C. = 506.6 Adjusted R-squared = .4683 Log likelihood = -480.5
---Estimated Standard
Variable Coefficient Error t-statistic P-value
---AGING_RATE 5.026 1.033 4.866 *** [.000]
PER_HOUSE_INCOME .6083E-03 .3457E-03 1.760 * [.081]
PER_HOUSE_INCOME2 -.1160E-08 .5917E-09 -1.960 * [.053]
UNEMPLOYMENT_RATE 8.290 2.944 2.816 *** [.006]
MEDICAL_AVAILABILITY 107.3 15.88 6.754 *** [.000]
HIGH_EDU_RATE 37.70 15.21 2.478 ** [.015]
HIGH_EDU_RATE2 -4.300 1.625 -2.647 *** [.009]
DIVORCE_RATE -116.2 35.41 -3.283 *** [.001]
DIVORCE_RATE2 20.11 6.570 3.061 *** [.003]
POPULATION_SIZE -.2061E-04 .6528E-05 -3.156 *** [.002]
POPULATION_SIZE2 .7687E-11 .1659E-11 4.633 *** [.000]
---Method of estimation = Ordinary Least Squares
Table 9: Regression for fitting improvement
Dependent variable: LNPER_CAPITA_MEDICAL_EXPENSES Current sample: 1 to 115
Number of observations: 115
Mean of dep. var. = 4.771 LM het. test = .7145 [.398]
Std. dev. of dep. var. = .1880
Sum of squared residuals = 1.403 Jarque-Bera test = 24.79 [.000]
Variance of residuals = .0136 Ramsey’s RESET2 = 1.944 [.166]
Std. error of regression = .1167 F (zero slopes) = 17.53 [.000]
R-squared = .6518 Schwarz B.I.C. = -61.71 Adjusted R-squared = .6146 Log likelihood = 90.18
---Estimated Standard
Variable Coefficient Error t-statistic P-value
---LNAGING_RATE .3159 .0900 3.510 *** [.001]
LNPER_HOUSE_INCOME 3.424 .3985 8.594 *** [.000]
LNPER_HOUSE_INCOME2 -.1450 .0181 -8.021 *** [.000]
LNUNEMPLOYMENT_RATE .2970 .0987 3.008 *** [.003]
LNMEDICAL_AVAILABILITY .3697 .0492 7.518 *** [.000]
LNHIGH_EDU_RATE 1.256 .7255 1.731 * [.086]
LNHIGH_EDU_RATE2 -.4271 .2477 -1.724 * [.088]
LNDIVORCE_RATE -1.366 .6958 -1.963 * [.052]
LNDIVORCE_RATE2 .6909 .3670 1.883 * [.063]
LNPOPULATION_SIZE -2.476 .3221 -7.685 *** [.000]
LNPOPULATION_SIZE2 .0927 .0121 7.659 *** [.000]
YILAN_DUMMY .2688 .0552 4.874 *** [.000]
---Method of estimation = Ordinary Least Squares
Table 10: Population by Three-Stage Age Group
Year Number in thousand Share in total:％
Western Taiwan 0-14 15-64 65+ Total 0-14 15-64 65+ Total 2010 99 3,634 17,046 2,486 23,166 15.7 73.6 10.7 100.0 2020 109 2,726 16,898 3,813 23,437 11.6 72.1 16.3 100.0 2030 119 2,503 15,115 5,683 23,301 10.7 64.9 24.4 100.0 2050 139 1,922 11,078 7,935 20,935 9.2 52.9 37.9 100.0 2060 149 1,775 9,219 7,843 18,837 9.4 48.9 41.6 100.0
Source: Table 17-3, Population by Three-Stage Age Group, Dependency Ratio and Ageing Index (Medium Variant), in the Population Projections for Taiwan: 2010-2060, (1999.9) The Council for Economic Planning and Develop-ment (CEPD).
Table 11: Simulation results of the future medical cost
a b c d e (=b×d)
Year Aging rate m = f (βX ) Index 1 Total Population Total m Unit ％ Thousand points 2010 =100 Thousand Billion points
2010 10.7 127.5 100.0 23,166 2,953
2020 16.3 124.5 97.7 23,437 2,918
2030 24.4 142.3 111.7 23,301 3,317
2050 37.9 161.6 126.8 20,935 3,383
2060 41.6 184.2 144.5 18,837 3,470
Source: Author’s calculation
Table 12: Simulation results on the burden of the future generation
e f g (=e/f) h
Year Total m 15-64 Per -workers Index 2
Unit Billion points Thousand Thousand points 2010 =100
2010 2,953 17,046 173.225 100.0