Forecasting Volatility in the Stock Market

Abstract

Utilizing an extensive array of technical indicators built upon historical stock price, volatility, and trading volume patterns, one may forecast the volatility of stock returns. Our out-of-sample findings show that adding technical variables to the autoregression benchmark can result in volatility projections that are far more accurate. We also compare the combined technical indicator predicting performance to that of the widely used economic indicators. This study forecasts stock market volatility and applies it to asset allocation, a common finance problem. We combine the drivers' predictive data to estimate market volatilities using machine learning and model averaging techniques. We confirm that the high-dimensional models outperform the typical volatility models in terms of prediction using a variety of evaluation techniques. This review's primary goal is to look at efficient GARCH models that are suggested for doing market returns and volatility analysis. This paper will talk about. predicting changes in stock market volatility.