The Determinants of Family Medical Expenditure：A Quantile Regression Approach

• Main Authors: Szu-Hsia Chen, 陳思遐
• Other Authors: Shih-Neng Chen
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/28297454071131876979

碩士 === 世新大學 === 經濟學研究所(含碩專班) === 93 === The expenditures on foods, dressings, and hosing etc. share a large portion of household income. As an economy develops and income increases, shares of above factors decline but shares of health care, entertainment, and communication increase. According to the 2002 Survey of Family Income and Expenditures (SFIE), as the aging of the population, the share of health care expenditure in GDP increases up to 12.3%. In traditional empirical analysis, one usually uses a regression approach to analyze the causal relationship between a dependent variable and a set of independent variables. In this study, we use a quantile regression method to assess the various marginal effects of an independent on the distribution of household health care expenditure in comparison with the OLS mean regression method. The household data used are the SFIE in 1992, 1997, and 2002. Empirical results show that quantile regression offers broader estimates of marginal effects than the OLS regression method. Family income and family medical expenditure have significantly positive relationship. The factors of family members, family mean age, number of kids (below 6 years old) and elders (above 65 years old), and residence region influence the portion of the highest medical expenditure quantile. In addition to economic factors, family traits have greater influences than interests and hobbies in the non-economic factors. The medical income elasticities of OLS estimates are 0.33 to 0.53 and of the quantile estimates are 0.209 to 0.539. This implies that family medical expenditure is a normal good and also a necessity.