An Evaluation of Alternative Models for Predicting Stock Volatility: Evidence from a Small Stock Market
Year: 1998
Type: preprint
Abstract: We examine the forecasting power of three recently proposed models for conditional stock volatility - an ARCH(9), a GARCH(1,2) and an AR(12)-model - with and without a seasonal component on a sample of monthly Swedish stock returns for the period 1977-1990. Our main results are the following: 1) the seasonal component adds forecasting power to all models, 2) the AR-model performs significantly better than both the ARCH- and the GARCH-model and 3) the AR-model performs at least as well as two benchmark forecasts - the implied volatility from stock index options and lagged actual volatility - despite the fact that these benchmark forecasts use a larger information set.
Source: SSRN Electronic Journal
Author: Per Frennberg
Institution: Lund University
Language: English
Subfield: Finance
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2299157815