Analysis and Applications of Autoregressive Moving Average Models with Stochastic Variance

Shelton Peiris, Ramprasad Bhar, David Allen


It is known that volatility plays a central role in financial modelling problems. This paper studies, in detail, a class of discrete time stochastic volatility (SV) models driven by ARMA models with innovations having a stochastic variances. The autocorrelation function of this class of models is derived and methods of identification of such processes are described. An example is added to illustrate the development of the theory over the standard methods.

Keywords: GARCH models, Volatility, Stochastic variance, Innovations, Heteroscedasticity, Random, Conditional expectation, Autocorrelation, Estimation.

AMS Subject Classification: Primary:62M10; Secondary:91B84.

This paper is available as a gzipped postscript (152kB) file and a PDF (208kB) file.
Tuesday, April 11, 2006