Volatility Prediction: A Comparison of the Stochastic Volatility, GARCH (1,1) and Egarch (1,1) Models.

Posted: 30 Apr 2001

See all articles by Harry M. Kat

Harry M. Kat

Independent

Ronald C. Heynen

Bank of America - Market Risk Management

Abstract

Future volatility is a key input for pricing and hedging derivatives and for quantitative investment strategies in general. There are many different approaches. This article investigates whether random walk, GARCH (1,1), EGARCH (1,1) and stochastic volatility models of return volatility behavior differ in their ability to predict the volatility of stock index and currency returns over horizons ranging from 2 to 100 trading days. We use close-to-close return data for 7 indices and 5 currencies over the period 1980-1992. The results show that the forecast performance of the different models depends on the specific asset class in question. For stock indices the best volatility predictions are generated by the stochastic volatility model. For currencies on the other hand, the best forecasts come from the GARCH (1,1) model.

JEL Classification: G00

Suggested Citation

Kat, Harry M. and Heynen, Ronald C., Volatility Prediction: A Comparison of the Stochastic Volatility, GARCH (1,1) and Egarch (1,1) Models.. Journal of Derivatives, Vol. 2, No. 2, 1994, Cass Business School Research Paper, Available at SSRN: https://ssrn.com/abstract=265008

Ronald C. Heynen

Bank of America - Market Risk Management ( email )

1 Alie Street
London E1 8DE
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
6,215
PlumX Metrics