No abstract

ISBN 978-80-7343-426-7 (Univerzita Karlova, Centrum pro ekonomický výzkum a doktorské studium) ISBN 978-80-7344-469-3 (Národohospodářský ústav AV ČR, v. v. i.) AbstractWe characterize formation of market beliefs in the oil market by providing a complete characterization of the market reaction to oil inventory surprises. We utilize the unique sequential nature of inventory announcements to identify inventory shocks. We estimate an AR-ARCH-MEM model of the joint dynamics of returns, return volatilities and trading volumes around the announcements using high frequency data on oil futures contracts. Our model (i) handles illiquidity of long maturity contracts by accounting for trading inactivity, (ii) captures time varying trading intensity, and (iii) allows for structural changes in the dynamics and responses to news over time. We show (i) uniform formation of expectations across oil futures contracts with different maturities, (ii) a strong negative relation between inventories surprises and re-turns, (iii) no effect on the term premium, which suggests that inventory shocks are always considered to be permanent, and (iv) di_erentiation in the reaction of volume by maturity. We demonstrate how our results can be used to test theories of oil price determination and contribute to the debate on the recent oil glut.

We show that ETF arbitrage distorts the market reaction to fundamental shocks. We conrm this hypothesis by creating a new measure of the intensity of arbitrage transactions at the individual stock level and using an event study analysis to estimate the market reaction to economic shocks. Our measure of the intensity of arbitrage is the probability of simultaneous trading of ETF shares with shares of underlying stocks estimated using high frequency data. Our approach is direct, and it accounts for statistical arbitrage, passive investment strategies, and netting of arbitrage positions over the day, which the existing measures cannot do. We conduct several empirical tests, including the use of a quasi-natural experiment, to conrm that our measure captures uctuations in the intensity of arbitrage transactions. We focus on oil shocks because they contain a large idiosyncratic component which facilitates identication of our mechanism and interpretation of the results. Oil shocks are identied using weekly oil inventory announcements.