THE SOURCES OF GARCH - EMPIRICAL-EVIDENCE FROM AN INTRADAY RETURNS MODEL INCORPORATING SYSTEMATIC AND UNIQUE RISKS

GARCH models propose that volatility is time varying and persistent. A s a parsimonious statistical description of the process driving return s in many financial markets, these models have been very successful. H owever, GARCH lacks a substantiated economic motivation. This paper pr ovides new and more definitive evidence on the mixture of distribution s hypothesis, a prominent potential economic explanation. The mixture hypothesis posits that autocorrelation in the time-varying rate of inf ormation arrival leads to the volatility dependencies captured by Garc h models. Previous investigations of this hypothesis have been inconcl usive. This paper makes use of a more theoretically appealing risk spe cification, more appropriate measures of the rate of information arriv al, and higher frequency data from the currency futures market. The re sults show that the mixture hypothesis explains substantial portions o f contracts' unique risks, but systematic GARCH remains.