[1]杨 毅,魏 艳,黄晓玲,等.基于指数平滑法的全国MRI配置量预测研究——四种非季节性模型比较[J].卫生经济研究,2020,(05):45-49.
　YANG Yi,WEI Yan,HUANG Xiao-ling,et al.Prediction of National MRI Configuration Based on Exponential Smoothing——Comparison of Four Non-seasonal Models[J].Journal Press of Health Economics Research,2020,(05):45-49.
点击复制

基于指数平滑法的全国MRI配置量预测研究
——四种非季节性模型比较
分享到：

卫生经济研究[ISSN:1004-7778/CN:33-1056/F]

卷:
期数:
2020年05期
页码:
45-49
栏目:
药械管理
出版日期:
2020-04-29

文章信息/Info

Title:
Prediction of National MRI Configuration Based on Exponential Smoothing
——Comparison of Four Non-seasonal Models
作者:
杨 毅1; 魏 艳1; 黄晓玲1; 张 晨1; 徐宁泽1; 程敬亮2; 陈英耀1
1. 复旦大学公共卫生学院，国家卫生健康委员会卫生技术评估重点实验室，上海 200032
2. 郑州大学第一附属医院，河南 郑州 450000
Author(s):
YANG Yi; WEI Yan; HUANG Xiao-ling; ZHANG Chen; XU Ning-ze; CHENG Jing-liang; CHEN Ying-yao
Key Laboratory of Health Technology Assessment of the Ministry of Health，School of Public Health，Fudan University，Shanghai 200032，China
关键词:
MRI; 指数平滑法; 非季节性模型; 预测
Keywords:
MRI; exponential smoothing; non-seasonal model; prediction
分类号:
R197.38
文献标志码:
A
摘要:
目的：对未来中短期全国MRI配置量进行预测。方法：基于存量数据，采用指数平滑法的四种非季节性模型，对未来全国MRI的配置量进行预测。结果：简单模型、Holt线性趋势模型、Brown线性趋势模型和阻尼趋势模型的R2值均大于0.95；后三者拟合后的数据残差序列不存在自相关（P>0.05），适宜采用指数平滑序列模型进行分析；Brown线性趋势模型的均方根误差和平均绝对误差百分比均为最小。结论：Brown线性趋势模型的拟合效果最佳，其预测结果显示，未来我国2020年、2023年和2025年MRI配置量分别为9 3
Abstract:
Objective To predict the national short-term MRI configuration. Methods Based on the stock data，four non-seasonal models using exponential smoothing were used to predict the future national MRI configuration. Results The R2 values of the simple model，Holt linear trend model，Brown linear trend model，and damped trend model were all greater than 0.95. The data residuals after the fitting of the latter three did not have autocorrelation (P>0.05)，so exponential smoothing was suitable. The serial model was used for analysis and the root mean square error and average absolute error percentage of the Brown linear trend model were at the smallest. Conclusion Brown's linear trend model had the best fitting effect，and it predicted that in the future China's MRI configuration in 2020，2023 and 2025 will be 9 332，11 750 and 13 362，respectively.

参考文献/References:

[1] 刘刚.ARIMA模型及其在麻疹发病率预测中的应用[J].数理医药学杂志,2011,24(4):379-382.
[2] Spaeder MC, Fackler JC. A multi-tiered time-series modelling approach to forecasting respiratory syncytial virus incidence at the local level[J]. Epidemiol Infect, 2012,140:602-607.
[3] Luz PM, Mendes BV, Codeco CT, et al. Time series analysis of dengue incidence in rio de janeiro, Brizal[J]. Am J Trop Med Hyg, 2008,79:933-939.
[4] Zhang X, Liu Y, Yang M, et al. Comparative study of four time series methods in forecasting typhoid fever incidence in China[J]. PloS one, 2013,8:e63116.
[5] 张磊,刘艳红.指数平滑法在预测深圳市宝安区肺结核病人发病人数的应用[J].实用预防医学,2014(8):911-913.
[6] 刘罗曼,时间序列分析中指数平滑法的应用[J].沈阳师范大学学报（自然科学版）,2009,27（4）:416-418.
[7] Jorgelina D’fana V, Shirley C, Luisa AP, et al. Incidence of tuberculosis at the local level: Marianao Municipality, Havana City, Cuba(1990-2000)[J].Rev Esp Salud Publica, 2003,77(2):221-231.
[8] 王昕,程小雯,房师松,等.指数平滑模型在流感样病例预测中的应用[J].中国热带医学,2011,11(8):938-939.
[9] Ratnasari S, Yuniaristanto, Zakaria R. Demand Forecasting with Five Parameter Exponential Smoothing[J]. IOP Conference Series: Materials Science and Engineering, 2019,495(1).
[10]唐广心, 张飞飞, 鲁苇葭, 等. 指数平滑法在麻疹发病率预测中的应用[J]. 实用预防医学, 2018,25(6):757-759.

相似文献/References:

[1]杨 毅,周星宇,魏 艳,等.MRI临床服务利用率评价研究[J].卫生经济研究,2020,(02):28.
　YANG Yi,ZHOU Xing-yu,WEI Yan,et al.Clinical Service Utilization Evaluation Study of MRI[J].Journal Press of Health Economics Research,2020,(05):28.

更新日期/Last Update: 2020-04-29