Three essays on the application of nonlinear regression models in health economics

In the first essay, we establish a theoretical model which illustrates that the diminished incentive to engage in physical activity can be related to the ex-ante moral hazard effects of the accessibility of medical care services. The Count Data model is used to examine the effects of health insurance coverage and the non-monetary cost of medical care services on the frequency of physical activity. The econometric evaluations based on the National Health Interview Survey (NHIS) in 1997 indicate that ex-ante moral hazard effects of health care services do exist. In particular, this study provides the first econometric evidence for the ex-ante moral hazard effects of access to health care services by altering the behavior of health investments.;The second essay attempts to examine if there is any substitutability between health habits and the demand for health care services among HMO members. Due to the fact that HMOs change the relative price of the physical activity and medical care services in terms of the two resources, time and money, individuals might substitute one input for the other in order to minimize the cost of production of the health stock. This hypothesis is supported by the empirical evidence from a newly proposed system equation regression model.;The third essay investigates the determinants of growth of aggregate health care expenditure (HCE) in OECD countries. The study departs from previous literature in that (1) it takes account of the slow adjustment of HCE and the endogeneity of explanatory variables by exploiting the panel structure of the dataset by Generalized Method of Moments estimation of a dynamic model and (2) it identifies other relevant determinants of HCE that have been overlooked in the existing literature such as the advancement of medical technologies and the degradation of the natural environment. One of the major findings is that per capita GDP is significantly positively related to HCE, but the magnitude suggests that health care is not a luxury good. Furthermore, HCE can also be explained by natural environmental quality and the medical technology developments in a degree.