Forecasting The Co-volatility Of Coffee Arabica And Crude Oil Prices A Multivariate Garch Approach With High Frequency Data

Forecasting the volatility dynamics of asset returns has been the subject of extensive research amongrnacademics, practitioners and portfolio managers. This thesis estimates a variety of multivariate GARCHrnmodels using weekly closing price (in USD/barrel) of Brent crude oil and weekly closing prices (inrnUSD/pound) of coffee Arabica, and compares the forecasting performance of these models based on a highrnfrequency intra-day data which allows for a more precise realized volatility measurement. The study usedrnweekly price data to explicitly model co-volatility, and employed high-frequency intra-day data to assessrnmodel forecasting performance. The analysis points to the conclusion that varying conditional correlationrn(VCC) model with Student’s t distributed innovation terms is the most accurate volatility forecasting modelrnin the context of our empirical setting. We recommend and encourage future researchers studying thernforecasting performance of MGARCH models to pay particular attention to the measurement of realizedrnvolatility, and employ high-frequency data whenever feasible.rnKeywords: commodity price co-volatility, conditional correlation, forecasting, multivariate GARCH

Get Full Work

Report copyright infringement or plagiarism