军交运费数据特征及预测模型研究

Data Characteristic and Prediction Model for Military Transportation Charge

针对军交运费数据的时序特性,提出基于差分自回归移动平均(ARIMA)模型的时间序列分析法。以某单位2009—2014年的军交运费值为样本数据,建立季节ARIMA模型,并预测2015年的军交运费。实验结果表明,所建立的模型充分考虑了军交运费的时序特性,具有较好的预测拟合精度,预测结果可为军交财务管理人员的经费预算工作提供科学依据。

To improve the correctness and utilization benefit of prediction value of military transportation charges, the paper puts forward a time series analysis method on the basis of autoregressive integrated moving average(ARIMA) model, and establishes a seasonal ARIMA model with a certain unit’s freight data in the period of time from 2009 to 2014, to forecast freight value in 2015. The experimental results show that the seasonal model has a good prediction fitting precision due to time series characteristics of charge data, and can provide scientific basis for financial managers to do budget work.