Date of Award




Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Economics

Content Description

1 online resource (xvi, 197 pages) : illustrations (some color)

Dissertation/Thesis Chair

Pinka Chatterji

Committee Members

Byoung Park, Baris Yoruk


Medical economics, Optometrists, Optometry, Drugs, Benzodiazepines

Subject Categories



My dissertation includes three main studies: the first study examines the seasonality of the birth weight and the underlying mechanisms for the seasonality; the second study documents the impact of weather conditions on the gestation period by using the duration analysis and on the birth weight by using the decomposition method and quantile regression; the third study analyzes the effects of density of immigrant population around place of residence on the residents’ health outcomes.In the first study, we evaluate a range of plausible mechanisms that may underlie the seasonality of the birth weight by using U.S. birth certificate records from 1989 to 2002. Although the seasonality of the birth weight is well documented in the epidemiology, the medical literature and, more recently, in the economics literature, not many studies attempt to explain the mechanisms underlying the seasonality. We therefore explore potential mechanisms including air pollution, maternal weight gains during pregnancy, number of prenatal care visits during pregnancy, unemployment rates, and weather conditions during pregnancy. We firstly replicate the seasonality of the birth weight that prior studies document by using the month of birth estimates. We then control the mechanisms to test how much the month of birth estimates change when we include each of the mechanisms separately. Our results indicate that air quality does not explain the seasonality of the birth weight because the month of birth estimates remain relatively the same after controlling the air quality. The weight gains and number of prenatal care visits during pregnancy reduce the estimates of the month of birth, but most of the estimates remain statistically and economically significant. Unemployment rates do not affect the seasonality of birth weight. On the other hand, controlling weather conditions during pregnancy reduces most of the month of birth estimates and they become statistically and economically insignificant except October and November month of births. Since the first dissertation chapter has demonstrated that solar radiation during pregnancy significantly affects the seasonality of birth weight, for further understanding, the second chapter attempts to investigate the impacts of solar radiation on the birth weight and the gestation period. We have three contributions to the prior literature of the weather conditions impacts on birth weights. Firstly, by applying the decomposition method through gestation periods, we analyze the impacts of solar radiation on birth weight and document direct and indirect impacts of solar radiation during pregnancy on birth weight. We employ the gestation period because gestation period is one of the main factors affecting the birth weight and solar radiation has a significant impact on the gestation period. Secondly, following the impact of solar radiation on gestation periods we find earlier, we extend our study to explore the impacts of solar radiation on gestation periods with the duration analysis to relax full-term birth assumption (gestation periods between 37 and 42 weeks) that prior studies had. Lastly, we identity the impacts of solar radiation on birth weight by employing quantile regressions as a means to focus on the distributional the impact of solar radiation during pregnancy. As mean regression targets on average, one cannot identify which groups of infants are more impacted by solar radiation. Our results evidence that solar radiation has much stronger negative impacts in the lower quantiles and solar radiation in fact has no impacts in the higher quantiles. Finally, using the individual-level data in the U.S. from the 2005 - 2012 Behavioral Risk Factor Surveillance System (BRFSS), linked with county-level data from the American Community Survey (ACS), we analyze the impacts of immigrant densities on residents’ health outcomes. Although ample literature has shown that neighborhoods have a substantial impact on residents’ health, prior studies demonstrate that the relationship between immigrant densities and health remains inconsistent. This study thus aims to look into how immigrant densities can affect residents’ physical and mental health and health behaviors. To account for the endogeneity, we use a two-stage least square (2SLS) estimation based on 1990 county immigrant density by each county in the U.S. Our results suggest that the higher immigrant density around place of residence reduces the risk of hypertension, heart attack, diabetes, stroke, depressive disorder, overweight, obesity, and cigarette smoking.

Included in

Economics Commons