Preliminary Version. 600 * 14530029 12-1-
Preliminary Version. 1 70 1 1997 9 1020 70 1 3 2 70 1 Manning et al. (1986) Rand Health Insurance Experiment 25 50 Manning et al. (1986) 3 Medicare Medicare Medicare Medigap plan Medigap plan Hurd and McGarry (1997) 1 70 2 3600-2-
Preliminary Version. Yoshida and Takagi (2002) 1 2 (professional norms) 3 2 3-3-
Preliminary Version. Evans (1974) Fuchs (1978 McGuire and Pauly (1991) Evans (1974) Fuchs (1978 4 Gruber and Owings (1996) 10 100 97 Yip (1998) 70 McGuire and Pauly (1991) Farley (1986) Jaegher and Jegers (2000) 4 McGuire (2000) -4-
Preliminary Version. McGuire and Pauly (1991) Farley (1986) Farley (1986) 1 1 10 2 3 4 2 122 8 4 9 8 17-5-
Preliminary Version. 60 65 70 70 69 70 70 70 Yoshida and Takagi (2002) 1 30 28 31 30 Two-part model Two-part model Sample Selection Models 5 5 Sample Selection Models Two-part models Jones -6-
Preliminary Version. 1 Sample Selection Models Two-part model 1 0 Z y hy ( z= 1) y f( y) I( y= 0) [ ] { [ ] } { [ Z ] [ Z ] } h y z f( y) = Pr Z = 0 Pr Z = 1 h( y z = 1) I( y 0) I( y= 0) I( y 0) I( y 0) = Pr = 0 Pr = 1 ( = 1) 2 2 y 70 70 2 random-effects panel probit median regression Median regression 1 median 1 quantile regression Quantile regression median regression Quantile regression Buchinsky (1998) 3 (2000) Mullahy (1998) -7-
Preliminary Version. 119 6 1 3.1 2 1965 16 84 27 8 4 9 8 17 70 70 8.5 0 12.6 9.7 70 10.4 11.0 25 50 75 3.7 7.3 12.1 4.0 7.3 12.6 4.0 7.3 12.4 6-8-
Preliminary Version. 3.3 7.1 13.3 4.6 7.7 12.9 4.3 7.6 13.0 25 50 75 70 70 10 30 8.97 17.89 10.27 8.74 1 70 1117 2217 1014 1013 119-9-
Preliminary Version. 72% 25% 22% 2% 4% 0.1% 1 7 (0, 0.05) (0.05, 0.10) 70 1 2 1 (0, 0.05) 70 (0.05, 0.10) 1 (0, 0.05) 70 70 (0.05, 0.10) 0.10 1 2 1 2 1 3.2 7 STATA 50 bandwidth plug-in STATA bandwidth -10-
Preliminary Version. 3 two-part Probit median regression 1 median median 2 Probit 2 Probit heterogeneity heterogeneity Probit 1 1 0 1 1 8 8 70 Newhouse et all. (1993) Long, Marquis and Rodgers (1998) RAND -11-
Preliminary Version. median regression 70 70 70 70 70 70 70 70 70 70 2 70 70 median median 3 median 70 70-12-
Preliminary Version. 3.3 1 1 1 1 4 70 70 17249 18236 1 719 809 90 70 100 70 100 150 70 100 2000 70 70 70 70 70 1 70 50 2 1 70 70 1 300 600-13-
Preliminary Version. 700 1100 500 700 1100 3.4 2 quantile regression 50% (median) 25% 75% 6 OLS 70 5 50% 70 70-14-
Preliminary Version. 70 381 70 88 163 70 76 138 98 70 70 110 70 10 1 81 70 81 2 25% 75% 54 70 25% 50% 1 75% 1 25% 50% 75% -15-
Preliminary Version. 119 900 119 1100 4 600 Farley (1986) professional norms -16-
Preliminary Version. 1 2 2 2 1 2 Buchinsky, M. (1998): Recent Advances in Quantile Regression Models: A Practical Guideline for Empirical Research, Journal of Human Resources, vol. 33, pp.88-126. Deb, P. and P. K. Trivedi (1997): Demand for Medical Care by the Elderly: A Finite Mixture Approach, Journal of Applied Econometrics, 12, pp.313-326. Hurd, M. D. and K. McGarry (1997): Medical Insurance and the Use of Health Care Services By the Elderly, Journal of Health Economics, 16, pp.129-154. Jones, A. M. (2000): Health Econometrics, Handbook of Health Economics, vol. 1A, eds. Culyer, A. J. and J. P. Newhouse, Elsevier, Amsterdam. Long, S. H., Marquis, M. S. and J. Rodgers (1998): Do People Shift Their Use of Health Services Over Time To Take Advantage Of Insurance?, Journal of Health Economics, 17, pp.105-115. Mullahy, J. (1998): Much ado about two: reconsidering retransformation and the two-part model in health econometrics, Journal of Health Economics, 17, pp.247-281. McGuire, T. G. and M. V. Pauly (1991): Physician Response to Fee Changes with Multiple Payers, Journal of Health Economics, 10, pp.385-410. Newhouse, J. P. et al. (1993): Free for All? Lessons from the RAND Health Insurance Experiment, Harvard Univ. Press, Cambridge, MA. -17-
69 70 7 6 5 4 3 2 1 0 0.05.1.15.2.25.3.35.4.45.5 69 70.0012.001.0008.0006.0004.0002 0 0 500 1000 1500 2000 2500 3000