Mixture Autoregressive Moving Average Model
混合自回归滑动平均模型———MARMA

A new mixture autoregressive moving average(MARMA) model is proposed for modeling nonlinear time series.The shape changing feature of conditional distributions makes the MARMA model capable of modeling time series with asymmetric,multimodal distribution,and conditional heteroscedasticity,and so on.The stationary conditions and autocorrelation function of the MARMA process are investigated. The Bayes information criterion(BIC) is used to select the MARMA model.The estimation of parameters is easily performed via expectation maximization(EM) algorithm.The model is applied to a financial data set and compared with other competing models.The MARMA model appears to capture features of the data better than other competing models do.