时间连续的马尔柯夫过程在汇率预测中的应用

为克服应用传统统计方法进行汇率预测时,因误选模型而导致的误差,提出了基于时间连续马尔柯夫过程的汇率预测方法。通过确定马尔柯夫过程的转移速度矩阵,建立汇率短期预测模型,并对其用拉普拉斯变换进行求解。将此模型应用于欧元/美元汇率的短期波动预测,实例表明,预测结果与实际观测值符合较好。

Forecasting of Exchange Rate Using the Continuous Time Markov Process

In order to get over the error following the improper model chosen to forecast the exchange rate by means of the traditional statistics, the continous tiroe Markov process is applied to forecast the short time exchange rate. By the aid of the transfer rate matrix, the model is established, that is sloved by the Laplace transform. Example shows that when the model is hpplied to forecast the short-time exchange rate of Euro to USD, the forecasted exchange rates have a good agreement with the actual values.