Application of garch models to estimate and predict financial volatility of daily stock returns in yahoo finance

Abstract

Investors and policymakers need to be aware of fluctuations in stock returns to manage portfolio adjustments and risk management decisions. Therefore, it is crucial to capture volatility, which is a measure of how strongly the price of a security clusters around the mean. The study used information from Yahoo Finance, a leading website of financial data and had applied the proposed methodology to one stock from each of the eleven industries in which they had separately represented the data. This project offers an approach to determine the best forecasting GARCH model among GARCH, EGARCH and GJR-GARCH that may be used to predict financial volatility of stocks. The study also used the Generalized Error Distribution, Students t Distribution, and Skewed Student t Distribution as error distributions in addition to the normal distribution. The identified models and error distributions that provided the significant parameters, were further forecasted using rolling window forecast and by relying on Root Mean Square error the best model had been selected.