基于卡尔曼滤波的期货价格期限结构模型

为准确对商品期货合约进行定价和预测,本文在短期-长期模型的基础上,提出以短期偏离、中期偏离和长期均衡为状态变量的三因素模型。本文根据状态变量的假设建立相关微分方程,并推导出模型的解,再运用卡尔曼滤波和极大似然法得到模型的参数和状态变量。最后,通过比较多种误差统计量证明,本文的短期-中期-长期模型的拟合与预测能力优于短期-长期模型。

Kalman Filter Based Term Structure Model of Futures Prices

In order to value and forecast the prices of commodity futures contracts,a three-factor model is put forward on the basis of short-term/long-term model,where short-term deviation,middle-term deviation and long-term equilibrium are state variables.Based on the assumptions on state variables,this paper constructs the differential equations,derives the solution from these equations,and then estimates the parameters and state variables by using Kalman filter and maximum likelihood method.At last,it is proved th...