Title: A mixed frequency stochastic volatility model for intraday stock market returns
Authors: Bastian Gribisch - University of Cologne (Germany) [presenting]
Jeremias Bekierman - University of Cologne (Germany)
Abstract: A mixed frequency stochastic volatility (MFSV) model is proposed for the dynamics of intraday asset return volatility. In order to account for long-memory we separate stochastic daily and intraday volatility patterns by introducing a long-run component that changes at daily frequency and a short-run component that captures the remaining intraday volatility dynamics. An additional component captures deterministic intraday patterns. The resulting non-linear state-space model is estimated in a single step using simulated maximum likelihood based on Efficient Importance Sampling (EIS). In addition to intraday volatility estimation and forecasting the model can be applied in order to forecast volatility at daily frequency incorporating intraday information on stock price fluctuations and daily realized volatility measures. We apply the model to intraday returns of five New York Stock Exchange traded stocks. The estimation results indicate distinct dynamic patterns for daily and intradaily volatility, where most of the volatility dynamics are explained by the daily volatility component. In-sample diagnostic tests and out-of-sample Value-at-Risk (VaR) forecasts show that already the very basic model specification successfully accounts for the strong persistence of intraday asset return volatility.