Date of Award

Spring 2022

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Barberis, Nicholas


In my dissertation, I dive into three specific areas in financial economics. Chapter 1 of my Ph.D. dissertation studies how the boom in off-exchange trading at the market close affects the close price discovery. In recent years, investment banks like Goldman Sachs have started a "guaranteed close" business where investors looking to buy or sell shares of a certain stock can get a guarantee from the bank to execute their orders at the close price set on the primary exchange. Using the TAQ data and a quasi-experimental shock from NYSE fee cut, we find that when the fraction of trades through "guaranteed close" increases, the informativeness of close price increases. We develop a model where investors choose which venue to trade in. A bank conducting "guaranteed close" business competes with the exchange on transaction fees, and gains profit from trading strategically utilizing the order flow information. The bank's trading activity concentrates the price-relevant information into the exchange. Consequently, the "guaranteed close" improves price discovery at the market close. Chapter 2 of my Ph.D. dissertation studies the long-term effects of experiencing high levels of job demands on the aging and mortality of CEOs. The estimation exploits variation in industry crises and takeover protection. First, we apply neural-network based ML techniques to assess visible signs of aging in pictures of CEOs. We estimate that exposure to a distress shock during the Great Recession increases CEOs' apparent age by one year over the next decade. Second, using hand-collected data on the dates of birth and death for 1,605 CEOs of large, publicly-listed U.S. firms, we estimate the resulting changes in mortality. The hazard estimates indicate that CEOs' lifespandecreases by 1.5 years in response to an industry-wide downturn, and increases by two years when insulated from market discipline via anti-takeover laws. Our findings imply significant health costs of managerial stress, also relative to known health risks. Chapter 3 of my Ph.D. dissertation provides an economically interpretable and easy-to-calculate approximation to optimal portfolio choice over the life cycle. The standard literature that solves the numerical optimal portfolio policy requires complicated backward induction, making it hard to apply for providing financial advice. Real-world financial advisors, on the other hand, tend to neglect the risky nature of human capital and offer advice that is not truly optimal. We bridge the gap by first using a reduced-form regression to predict discount rates of future incomes over an agent's life. Our prediction method achieves an R-squared of more than 90% over a wide range of simulations. Furthermore, by plugging the discount rates we predict into Merton (1969) formula, we obtain an approximate solution that has an average difference within 2% when compared to optimal solution solved through backward induction.