by
Mark Manfredo
Morrison School of Agribusiness and Resource Management
Arizona State University
Raymond M. Leuthold
Department of Agricultural and Consumer Economics
University of Illinois at Urbana-Champaign
and
Scott H. Irwin
Department of Agricultural and Consumer Economics
University of Illinois at Urbana-Champaign
Abstract
Considerable research effort has focused on the forecasting of asset
return volatility. Debate in this area centers around the performance of
time series models, in particular GARCH, relative to implied volatility
from observed option premiums. Existing literature suggests that the performance
of any volatility forecast is sensitive to both the data and forecast horizon
of interest. This paper rigorously examines the performance of several
alternative volatility forecasts for fed cattle, feeder cattle, and corn
cash price returns. Forecasts include time series, implied volatility,
and composite specifications. The results provide considerable insight
into the performance of these alternative volatility forecasting procedures
over a range of relevant forecast horizons. The evidence suggests that
composite methods be used when both time series and implied volatilities
are available. Insight is also gained into the performance of procedures
used for scaling one-period volatility forecasts to longer horizons. However,
consistent with the existing volatility forecasting literature, this research
confirms the difficulty in finding a "best" volatility forecasting method
across alternative data sets and horizons.