Minimally cross-entropic conditional density : a generalization of the GARCH model

The stylized fact of time-varying volatility in financial series is commonly accepted amongst scholars as well as practitioners. The GARCH model has been exceptionally successful in this area. Our approach, the minimally cross-entropic conditional density (MCECD) model, is a generalization of GARCH(1,1) which can cope with conditional skewness and kurtosis. It is so-named because the parameter updating method is based on cross-entropy minimization rather than autoregression.