Essays in asset pricing and market imperfections

Thesis (Ph. D.)--Massachusetts Institute of Technology, Sloan School of Management, 2010.Cataloged from PDF version of thesis.Includes bibliographical references (p. 171-176).The first part of the thesis studies the impact of liquidity crashes on asset prices. In financial markets, liquidity could have large downward jumps. The thesis proposes a dynamic model where investors face the risk of potential liquidity crises. We find that investors choose optimal portfolios not only to hedge the risk of asset fundamentals, but also to hedge the risk of potential liquidity crashes. The potentially illiquid assets tend to have a lower price, a higher volatility, and a lower volume turn-over. Liquidity hedging could induce high return premium and asset returns could have excess volatility over the fundamentals. The risk of potential liquidity crises will also generate rich patterns in return dynamics and the expected asset returns could be driven by risks that are not systematic. The second part of the thesis analyzes the effect of illiquidity on the extreme risk of hedge funds. Hedge funds' returns often exhibit positive autocorrelations, which suggests illiquidity in their asset holdings. In this part, using a data set containing monthly returns of over 5,600 hedge funds, I study how illiquidity affects the extreme risk of hedge funds. I use MA(q) processes to model hedge funds' returns and use smoothing coefficients as proxies for liquidity. The tail risks are estimated using the extreme value theory and the generalized Pareto distribution. We find that illiquidity in general has a negative impact on the tail risk of hedge funds' returns. In particular, the true Value-at-Risk (VaR) of hedge funds could be much higher when illiquidity is taken into consideration.(cont.) The third part of the thesis studies asset pricing under heterogeneous information. In an asset market where agents have heterogeneous information, asset prices not only depend their expectations of the true fundamentals but also depend on their expectations of the expectations of others. Iterations of such expectations lead to the so-called "infinite regress" problem, which makes the analysis of asset pricing under heterogeneous information challenging. In this part, we solve the infinite-regress problem in a simple economic setting under a fairly general information structure. This allows us to examine how different forms of information heterogeneity impacts the behavior of asset prices, their return dynamics, trading volume as well as agents' welfare.by Weiyang Qiu.Ph.D