Forecasting implied volatility
Keywords
Loading...
Authors
Issue Date
2024-07-30
Language
en
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This study tests methods of forecasting implied volatility. Standard Black-Scholes option pricing assumes that volatility is constant over time, but numerous empirical studies have verified that the conditional volatility of stock prices is clustered and noisy. Market efficiency arguments lead to the expectation that investors would include these volatility patterns into option prices if they could, more sophisticated option valuation models such as Heston (1993) allow them to do exactly that. GARCH models (as proposed by Bollerslev (1986)) allow for the modelling and forecasting of conditional volatility which is potentially clustered. This leads to the question whether future implied volatility can be forecasted using volatility forecasts made with GARCH models.
Results show that GARCH forecasts are statistically significant predictors of implied volatility, but that these do not outperform ARMA forecasting. This leads to the conclusion that GARCH is actually not the best way of forecasting implied volatility. Moreover, it seems that the market also considers additional factors not investigated here.
Description
Citation
Supervisor
Faculty
Faculteit der Managementwetenschappen