Denghui Chen

Date of Graduation


Document Type


Degree Type



College of Business and Economics



Committee Chair

Victor Chow

Committee Co-Chair

Alexander Kurov

Committee Member

Eric Olson

Committee Member

Gulnara Zaynutdinova


This dissertation explores issues regarding the effect of investor risk aversion and sentiment on financial markets. It is widely considered that a risk averse investor requires risk premium to hold risky assets, and the required risk premium is proportional to the riskiness of the underlying assets. From the myopic loss aversion perspective, investors, who get utility by frequently evaluating their portfolios and are more sensitive to losses, will require a higher risk premium as compensation for the fear of a major drop in financial wealth. Recent literature has shown that large tail jumps are often contributed to such major drop in financial markets. Rare events, which often accompanied with tail jumps, have a more drastic impact on risk averse investor, and the compensation for rare events accounts for a large fraction of the equity risk premium. However, there is no theoretical framework that has been developed to separate the risk aversion component of rare events from daily volatility.;In this work, I continue to argue the importance of the rare events have different impact on investors decision regarding to equity risk premium. Rational investor risk aversion should spike and require higher risk premium to compensate for higher risk when rare events happen. I develop a theoretical framework to decompose the risk aversion component with rare events from the part associated with daily volatility and prove that they have varying impact on risk premium. Specifically, I first extend the jump-diffusion model incorporated with disaster models and develop a theoretical framework to decompose the risk aversion component of rare events from frequent events. Then I attempt to use monetary surprises as empirical application of the theoretical framework. Bernanke and Kuttner (2005) show that the effect of monetary surprises on expected excess returns can be related to the impact of monetary policy on investor risk aversion or the riskiness of stocks. Finally I test how market response to macroeconomic news surprises conditional on investor sentiment. The results are arranged in the following order.;Chapter 1 of the dissertation makes the argument that investors should treat rare events differently from frequent events. Intuitively, a rare event (disaster) reduces the fundamental value of a stock by a time-varying amount. A rise in disaster probability lowers the expected rate of return on equity, and it also motivates investors to shift toward the risk-free asset or buy deep out-of-the-money puts. Comparing to previous models, I extend the general jump-diffusion model by decomposing investor risk aversion towards frequent and rare events separately. I show that investor treats frequent volatility (quadratic variations) and rare events (tail variations) differently, and decompose representative agent's risk aversion into volatility risk aversion and tail risk aversion. The model implicates that tail risk aversion can be considered as a tractable way to model changes in expectation of rare events and equity risk premium.;In Chapter 2 of the dissertation, I apply the model implication by using monetary surprises and decomposed risk premium from the VIX. Monetary surprises are not only likely to influence stock prices, but also to affect the degree of uncertainty and risk aversion faced by investors. This paper examines the interdependence of stock market, tail risk aversion, and monetary shocks across conventional and unconventional monetary policy periods. The empirical sample is from January 1998 to October 2014. During conventional monetary policy period, I find that there is contemporaneous response among stock market, time-varying risk aversion, and monetary policy under a VAR analysis. Generally, both volatility and tail risk aversion components respond negatively to a positive stock market and monetary policy shock. It indicates that a bullish stock market decreases investor risk aversion, and a positive monetary shock also decreases investor risk aversion as increasing interest rate is considered a good indicator of the economy. During unconventional times, we find that a surprise decline in the expected short-term rate leads to an increase in stock prices and a mixed effect on tail risk aversion. A plausible explanation is that investors believe a surprise drop in expected short-term rate reflects a fast deteriorating economic outlook during unconventional monetary policy period.;Chapter 3 of the dissertation is co-authored project. In this chapter, we consider the effect of investor sentiment from market microstructure perspective. We consider investor sentiment arises when noise traders actively trade on pseudo-signals (Shiller, 1984; Lee, 2001), and investigate the effect of investor sentiment on market reaction to macroeconomic news surprises. Noise traders are considered to respond to new information sub-optimally, and smart-money investors respond to news about fundamental value rapidly and optimally. We utilize both simultaneous news release analysis and normalized estimation proposed by Swanson and Williams (2014). (Abstract shortened by ProQuest.).