基于ARIMA和Holt指数平滑模型的山东省药品集中采购价格指数研究

Centralized Drug Purchase Price Index of Shandong Province Based on ARIMA and Holt Index Smoothing Model

本文实现了对“十三五”期间山东省药品集中采购价格的动态分析和预测。利用采购平台提供的数据编制链式拉氏价格指数、帕氏价格指数,将价格指数作为对药品集中采购价格预测分析的重要指标依据,以此进行量化评价。结合ARIMA和Holt指数平滑模型,对已有的药品价格指数数据进行预测,两个模型的平均绝对误差分别为0.3611和0.3723,表明模型能够有效拟合序列的整体趋势,以及提取价格指数序列动态及持续特征。通过对价格指数进行趋势预测以评估政府及相关部门稳控药价等政策的效果,为政府和平台管理者提供科学的决策依据。

This paper realizes the dynamic analysis and forecast of the centralized drug purchase prices in Shandong Province during the 13th Five-Year Plan period.By using the data provided by the procurement platform,the chain Laspeyres price index and Paasche price index were compiled,and the price index was used as an important index basis for the prediction and analysis of centralized drug purchase price,so as to carry out quantitative evaluation.Combining the ARIMA with Holt exponential smoothing model to predict the existing drug price index data,the average absolute errors of the two models are 0.3611 and 0.3723,respectively,which indicates that the models effectively fit the overall trend of the sequence,and extract well the dynamic and persistent characteristics of price index series.With the trend prediction of the price index,the implementation effect of drug price stabilization policies of the government and relevant departments are evaluated,and scientific decision-making basis is provided for the government and platform managers.