Show simple item record

dc.contributor.advisorZivot, Ericen_US
dc.contributor.authorLee, Jee Youngen_US
dc.date.accessioned2012-09-13T17:38:24Z
dc.date.available2012-09-13T17:38:24Z
dc.date.issued2012-09-13
dc.date.submitted2012en_US
dc.identifier.otherLee_washington_0250E_10513.pdfen_US
dc.identifier.urihttp://hdl.handle.net/1773/20862
dc.descriptionThesis (Ph.D.)--University of Washington, 2012en_US
dc.description.abstractThis dissertation studies the U.S. stock market. The first chapter explores a four-moment CAPM under regime switching which incorporates the risk premia for skewness and kurtosis. As expected, estimates of risk premia for covariance, co-skewness and co-kurtosis risks are different across regimes. By allowing time-varying (regime-specific) volatility correlations among asset specific innovations it captures strong volatility correlations in a crash state. As the market evolves to a more bullish state the volatility correlations weaken. Large changes in skewness and kurtosis are linked to regime switching. Optimal weights within a portfolio of small-caps, large-caps, and a risk-free asset are different across regimes when skewness and kurtosis preferences are considered. The second chapter investigates price discovery and information revealing patterns during the two recent volatile U.S. stock market periods (Tech bust 2000 and credit crisis 2008). In volatile markets, a large-cap stock (MSFT) reveals a considerable amount of private information through trades before the open and during trading hours while a small-cap stock (OPNT) shows little trade-correlated information over the trading day. This is contrary to a typical expectation that small firms' trades are more informative. Information interactions between two stocks are most active during the first-half and the last-half of trading hours while dormant during mid-day trading. This implies information revealing patterns could be different in volatile markets. The third chapter provides a framework to improve risk-adjusted returns in tactical asset allocation. Tactical asset allocation involves judgments of the future asset returns in a portfolio and thus it is important to identify the market turns. With this motivation, a Markov switching model is applied to the spread returns between small-caps and large-caps to specify market states. Next, a dynamic ordered probit model is adopted to estimate market bullishness (latent variable) and forecasts of the latent variable are used to get tactical tilts between two opposing assets. To get better forecasts of asset returns, the higher moments of the regime switching model are incorporated since changes in skewness and kurtosis reflect changing market conditions. Finally, it is observed that higher moments can provide better downside protections in tactical asset allocation.en_US
dc.format.mimetypeapplication/pdfen_US
dc.language.isoen_USen_US
dc.rightsCopyright is held by the individual authors.en_US
dc.subjectHigher Moments CAPM; Information and Market Efficiency; Skewness and Kurtosis; Tactical Asset Allocationen_US
dc.subject.otherFinanceen_US
dc.subject.otherEconomicsen_US
dc.subject.otherEconomicsen_US
dc.titleEssays on Financial Econometricsen_US
dc.typeThesisen_US
dc.embargo.termsNo embargoen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record