节点文献

基于LSTM和ARIMA模型的山东省抗肿瘤药物价格水平预测研究

Price prediction of antitumor drugs in Shandong Province based on LSTM and ARIMA models

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 曹晓宇徐波张慧凤魏新江

【Author】 Cao Xiaoyu;Xu Bo;Zhang Huifeng;Wei Xinjiang;School of Mathematics and Statistics Science, Ludong University;Shandong Public Resources Trading Center;

【通讯作者】 魏新江;

【机构】 鲁东大学数学与统计科学学院山东省公共资源交易中心

【摘要】 目的 评估2016年1月至2021年4月期间山东省抗肿瘤药物的价格水平,科学预测抗肿瘤药物的价格走势。方法 基于脱敏处理的2016年1月至2021年4月期间山东省抗肿瘤药物采购数据,编制抗肿瘤药物价格指数,利用ARIMA模型和LSTM模型对抗肿瘤药物价格走势进行预测对比分析。结果 山东省抗肿瘤药物的拉式、帕式、费式价格指数呈下降趋势,ARIMA和LSTM模型的MSE分别为0.60%和1.85%,ARIMA模型比LSTM模型更适合进行预测。利用ARIMA模型预测抗肿瘤药物未来8个月价格趋势,整体呈现递减趋势。结论 山东省药品集中采购政策在一定程度上能够降低抗肿瘤药物价格,但抗肿瘤药物价格有异常波动情况。医保卫健等相关部门应从带量采购、国家医保谈判等方面入手确保抗肿瘤药物价格在合理范围。

【Abstract】 Objective To evaluate the price level of antineoplastic drugs in Shandong Province from January 2016 to April 2021, and to scientifically predict the price trend of antineoplastic drugs.Methods Based on the purchase data of desensitized anti-tumor drugs from January 2016 to April 2021 in Shandong Province, the anti-tumor drug price index was compiled, and the ARIMA model and the LSTM model were used to predict and compare the price trend of anti-tumor drugs.Results The Laspeyres, Paasche and Fisher price indices of antitumor drugs in Shandong Province showed a decreasing trend, and the MSE of the ARIMA and LSTM models were 0.60% and 1.85%, respectively. The ARIMA model was more suitable for prediction than the LSTM model. ARIMA model was used to predict the price trend of antitumor drugs in the next 8 months, and the overall trend was decreasing.Conclusion Centralized drug procurement policy in Shandong Province can reduce the price of antitumor drugs to a certain extent, but the price of antitumor drugs fluctuates abnormally. Relevant departments such as medical security and health care should ensure the price of antitumor drugs in a reasonable range from the aspects of bulk purchase and national insurance negotiation.

【基金】 国家自然科学基金(61973149);山东省自然科学基金重点项目(ZR2020KF029);山东省社科规划研究项目(21CSDJ20)
  • 【文献出处】 中国医院统计 ,Chinese Journal of Hospital Statistics , 编辑部邮箱 ,2022年06期
  • 【分类号】R956
  • 【下载频次】12
节点文献中: 

本文链接的文献网络图示:

本文的引文网络