Repository Article

·2012 OPEN ACCESS

Support Vector Machine GARCH and Neural Network GARCH Models in Modeling Conditional Volatility: An Application to Turkish Financial Markets

Melike Bildirici YTU , Özgür Ömer Ersin

SSRN Electronic Journal

Abstract

The Turkish version of this paper can be found at: <a href="http://ssrn.com/abstract=2222071">http://ssrn.com/abstract=2222071</a> The study aims to investigate linear GARCH, fractionally integrated FI-GARCH and Asymmetric Power APGARCH models and their nonlinear counterparts based on Support Vector Regression (SVR) and Neural Network (NN) models. GARCH family models are extended to NN-GARCH architecture of Donaldson and Kamstra (1997) to various NN-GARCH family models (Bildirici and Ersin, 2009) such as NN-APGARCH model. The study aims to introduce a class of extended NN-GARCH and SVR-GARCH family of models with nonlinear augmentations in modeling both the conditional mean and variance. The SVR-GARCH, SVR-APGARCH and SVR-FIAPGARCH and their Multi-Layer Perceptron architecture based counterparts, MLP-GARCH, MLP-APGARCH and MLP-FIAPGARCH are evaluated. An application to daily returns in Istanbul ISE100 stock index is provided. Results suggest that volatility clustering, asymmetry and nonlinearity characteristics are modeled more efficiently with the models possessing neural network architectures.

Keywords

Autoregressive conditional heteroskedasticity Volatility (finance) Turkish Artificial neural network Econometrics Finance Financial economics Economics Computer science Artificial intelligence

Subject Areas

Financial Risk and Volatility Modeling ·Finance ·Social Sciences
Stock Market Forecasting Methods ·Management Science and Operations Research ·Social Sciences
Market Dynamics and Volatility ·Economics and Econometrics ·Social Sciences

Citations by Year

OpenAlex SDG Match

SDGs auto-classified by OpenAlex (score ≥ 0.4 shown).

Industry, innovation and infrastructure 52%