农产品市场价格短期预测方法的选择及应用——以鲜奶零售价格为例

Choice and Application of Short-term Forecast Method for Agricultural Products Price——Taking Fresh Milk Retail Price as Example

本文在概述时间序列模型种类及特点的基础上,以我国鲜奶零售价格为例,示范了进行农产品市场价格短期预测时选择合适时间序列模型的筛选过程。通过平稳性、季节性、趋势性以及异方差等一系列检验后,本研究最终选择双指数平滑、Holt-W inters无季节性模型和ARCH模型共3种方法对我国鲜奶零售价格短期预测进行了应用模拟,结果显示,ARCH模型预测结果精确度最高,Holt-W inters无季节性模型稳定性最好。

Based on summarizing the kinds and characteristics of the time series models and taking our country fresh milk retail price as example, the process of choosing the right time series models for short - term forecast of agricultural products price was demonstrated. Checked by stationarity, seasonal characteristic, trend and heteroskedasticity etc. , the exponential doubles smoothing, Holt- Winters non -seasonal smoothing and the ARCH model were chosen for analyzing the fresh milk retail price. The results indicated that the forecast precision of ARCH model was the best, and the forecast stability of Holt - Winters non - seasonal smoothing was the best.