Gaussian Scale Mixture Model for Estimating Volatility as a Function of Economic Factor

In this paper the scale mixture of Gaussian distribution is used to model the stock return data in financial market. There are many volatility models and forecasting methods. Some of the models are Historical volatility models, Implied volatility models, Autoregressive Conditional Heteroskedasticity models, models based on Artificial Neural Network. All these models are direct models. In these models the influence of economic factors like price level uncertainty, riskless rate of interest, the equity risk premium and the ratio of expected profit to expected revenue for the economy are not taken into account. Here the volatility parameter ‘ ’ is treated as a function of an economic factor. The main economic factor considered is the ratio of expected profit to expected revenue. Economic ratio is assumed to follow the exponential distribution. The resultant distribution is fitted to Dow Jones Industrial Average (DJIA) data by estimating the parameters. It is observed that this mixture distribution is a better fit than the GARCH fit. JEL Classification Code: C1