Essays in empirical asset pricing
Faculty Advisor: Andrew Ang
This dissertation is concerned with empirical evidence on the pricing of risky assets. The first chapter asks whether the surge in risk spreads during the recent financial crisis owes to credit or liquidity problems. To address this question, I form new credit and liquidity risk measures and then use these to decompose interest rate spreads into credit and liquidity components. While credit and liquidity are both important in explaining interest rate spreads, I find that liquidity effects explain a much larger share of the rise in risk spreads than had previously been found by other researchers. The second chapter investigates the effect of futures market participant positioning announcements by the Commodity and Futures Trading Commission on equity futures prices. Using high-frequency intradaily data, I find that these announcements have a significant effect: news that speculators have net long futures positions leads equity prices to rise. This is consistent with a view that speculators have an information advantage. The final chapter, which is joint work with Andrew Ang and Jun Liu, revisits the estimation of factor risk premia: a classic problem in empirical asset pricing. The existing literature almost invariably uses portfolios as test assets on the grounds that their betas are estimated more precisely. This chapter uses theoretical, empirical, and Monte-Carlo evidence to compare the properties of factor risk premium estimates when the test assets are portfolios versus individual stocks. Contrary to the conventional wisdom, the most efficient estimates are obtained when using individual stocks.