Generalized autoregressive conditional heteroscedastic modeling in finance: Inferences from stock prices.

摘要

Recently, there has been a growing interest in using Generalized Autoregressive Conditional Heteroscedastic (GARCH) models in the finance discipline. The discussion will begin with an overview of the GARCH(P,Q) processes and its properties, with an emphasis on the normal GARCH(1,1) model. The highlight of this thesis lies in the two numerical examples investigating the usage and performance of several GARCH models. The first example features the fitting of six GARCH models to the returns of the Standard & Poor's 100 daily index (January 1991-December 2000). The second example explores three GARCH models utilizing the risk-neutral valuation of call options via Monte Carlo simulation for the Standard & Poor's 100 daily index (January 1981-December 1993) and the Standard & Poor's 500 weekly index (January 1991-December 2000).