基于指数平滑法与马尔科夫链的汽车拥有量预测模型

The Prediction Model of Car Ownership Based on Exponential Smoothing and Markov Chain

民用汽车拥有量预测是城市交通规划、交通管理与控制的基础。为准确预测民用汽车拥有量的发展趋势,首先运用指数平滑法建立了汽车拥有量预测模型并利用实际数据对模型的准确性进行了验证;然后采用质心聚类法对相对误差进行状态划分并确定状态区间;最后利用马尔科夫链对初始预测模型进行修正,进一步提高了模型的预测精度。预测结果表明,未来几年全国每千人汽车拥有量仍将持续快速增长,建议有关部门及时采取措施,以避免交通拥堵、环境、能源等问题的加重。

The prediction to the civil vehicles ownership is the basis of urban traffic planning,traffic management and controlling.In order to accurately predict the development of civil car ownership,the exponential smoothing method is used to establish the prediction model firstly,and the accuracy of the model is verified based on actual data.Then the centroid clustering method is used to divide the relative errors into different states and determine the state intervals.Finally,the Markov chain is used to modify the initial prediction model,which can further improve the prediction accuracy of the model.The forecast results show that the number of vehicles per thousand people in China will continue to grow rapidly in the next few years.It is recommended that relevant departments take measures in time to avoid aggravation of traffic congestion,environmental and energetic problems.