Sector Volatility Prediction Performance Using GARCH Models and Artificial Neural Networks

TL;DR

This study compares GARCH and ANN models for stock volatility prediction across different market sectors, finding that ANNs excel in low-volatility cases while GARCH models perform better with medium and high volatility assets.

Contribution

It provides a comparative analysis of ANN and GARCH models across various volatility profiles, guiding model selection for different asset types.

Findings

ANN models outperform GARCH for low-volatility assets

GARCH models are more accurate for medium and high-volatility assets

Model choice depends on the asset's volatility profile

Abstract

Recently artificial neural networks (ANNs) have seen success in volatility prediction, but the literature is divided on where an ANN should be used rather than the common GARCH model. The purpose of this study is to compare the volatility prediction performance of ANN and GARCH models when applied to stocks with low, medium, and high volatility profiles. This approach intends to identify which model should be used for each case. The volatility profiles comprise of five sectors that cover all stocks in the U.S stock market from 2005 to 2020. Three GARCH specifications and three ANN architectures are examined for each sector, where the most adequate model is chosen to move on to forecasting. The results indicate that the ANN model should be used for predicting volatility of assets with low volatility profiles, and GARCH models should be used when predicting volatility of medium and high volatility assets.