焦炭价格预测研究具有重要的理论和实践意义,本文利用卡尔曼滤波算法对焦炭价格进行预测研究。建立状态空间模型时,选取焦炭价格作为唯一的状态变量,通过每一时刻变量观测值与预测值形成的新息,不断更新和迭代,以寻求最优估测值。实证分析表明,该算法对焦炭价格的跟踪和预测效果较好。 Coke price forecasting research has important theoretical and practical significance. In this paper, the use of Kalman filter algorithm to predict coke prices. When establishing the state space model, the price of coke is selected as the only state variable, and the new information formed by the observed value of the variable and the predicted value at each moment is continuously updated and iterated to find the optimal estimation value. The empirical analysis shows that the algorithm has a good effect on tracking and forecasting coke prices.