| This thesis contains two different topics that investigate issues in time series econometrics and applied health economics.;Both the second and third chapters study the obesity epidemic in Canada in recent years. Chapter 2 focuses on changes in obesity prevalence of Canadian adults, while chapter 3 focuses on their BMI shift pattern among different BMI categories.;By applying the quantile regression to three health surveys conducted from 1978 to 2004, chapter 2 explores the effects of some widely used demographic, geographic, and socioeconomic factors, as well as lifestyle, on the body mass index (BMI). The results of this study show that, changes of BMI that are attributable to these factors differs at different points of the BMI distribution, and the importance of any given factor changes over time. The epidemic of obesity in recent years is more closely in related to lifestyle and socioeconomic factors than demographic factors. By applying the Markov chain analysis to the BMI for individuals from the longitudinal National Population Healthy Survey (NPHS), chapter 3 explores how the pattern of weight of Canadian individuals has shifted among six categories of BMI during the 1994/95--2006/07 period. Two policy implications are suggested by this study: first, the effort to prevent the occurrence of new obese cases seems to be much more effective than the effort to reduce the number of existing obese people, and second, health policies for reducing obesity should focus more on the physically inactive people and provide incentives for them to be more physically active.;The first chapter (joint work with Nikolay Gospodinov) proposes a bootstrap unit root test in models with GARCH(1,1) errors and establishes its asymptotic validity under mild moment and distributional restrictions. While the proposed bootstrap test for a unit root shares the power enhancing properties of its asymptotic counterpart (Ling and Li, 2003), it corrects the substantial size distortions of the asymptotic test that occur for strongly heteroskedastic processes. The simulation results demonstrate the excellent finite-sample properties of the bootstrap unit root test for a wide range of GARCH specifications. |
