A MAXIMUM-LIKELIHOOD APPROACH FOR NON-GAUSSIAN STOCHASTIC VOLATILITY MODELS

A maximum likelihood approach for the analysis of stochastic volatilit y models is developed. The method uses a recursive numerical integrati on procedure that directly calculates the marginal Likelihood. Only co nventional integration techniques are used, making this approach both flexible and simple. Experimentation shows that the method matches the performance of the best estimation tools currently in use. New stocha stic volatility models are introduced and estimated. The model that be st fits recent stock-index data is characterized by a highly non-Gauss ian stochastic volatility innovation distribution. This model dominate s a model that includes an autoregressive conditional heteroscedastic effect in the stochastic volatility process and a model that includes a stochastic volatility effect in the conditional mean.