A Business Intelligent Model for Market Risk Measurement

In this study, we propose a business intelligent model integrating econometric models, i.e. ARMA, GARCH, and ANN models for VaR estimation. The business intelligent model achieves better efficiency in input variables selecting because they are selected and newly created by time series models. Repetitive trial error process could be effectively eliminated to one time series process. On the other hand, the performance of traditional time series models could be further enhanced by the forecasting power of ANN models. Empirical study shows that the business intelligent model can improve the predictive power in the framework of both accuracy and reliability.