Empirical analysis: Application of specific GARCH models in examining stock market volatility

• Main Authors: Canonizado, Adrianne Nicole J., Chua, Charles Lawrence L., Go, Jon Pryce Y., Yu, Jackie C.
• Format: text
• Language: English
• Published: Animo Repository 2021
• Subjects: Stock exchanges—Philippines
• Online Access: https://animorepository.dlsu.edu.ph/etdb_finman/38; https://animorepository.dlsu.edu.ph/cgi/viewcontent.cgi?article=1001&context=etdb_finman
• Institution: De La Salle University

One permanent characteristic of every stock market is volatility. Examining and forecasting stock market volatility is important for several stakeholders including the traders, government, future researchers. Despite this, little to no studies have been conducted to establish which among the widely-used methodologies in predicting stock market volatility were the most appropriate for the Philippine stock market characteristics. The primary purpose of the research study is to compare five GARCH-family models with regards to their capabilities in modeling seven different indices in the Philippine Stock Exchange, namely: financial, industrial, holding firms, property, services, mining and oil, and the PSEi. The study also seeks to determine which among these models outperforms the others per index. In doing so, daily stock prices during the period of 2010 to 2019 were obtained from the PSE web portal, and the log of the returns are taken as inputs to the models in this study for observation. The findings of the study suggest that the GJR-GARCH model outperformed the other GARCH models in the case of the financial, property, and services indices. Furthermore, the TGARCH is superior in the case of the industrial, holding firms, and the PSEi. Lastly, the EGARCH model was found to have outperformed the other models in the case of the mining and oil index.