Chin J Schisto Control ›› 2020, Vol. 32 ›› Issue (3): 236-.

Previous Articles     Next Articles

Application of the exponential smoothing model and ARIMA model in prediction of the endemic situation of schistosomiasis in Hunan Province

ZHOU Jie, REN Guang-Hui, HE Hong-Bin, HOU Xun-Ya, DENG Wei-Cheng*   

  1. Hunan Institute of Parasitic Diseases, WHO Collaborating Center on Schistosomiasis Control in Lake Regions, Hunan Key Laboratory of Immunology and Transmission Control of Schistosomiasis, National Key Clinical Specialty, Yueyang 414000, China
  • Online:2020-05-16 Published:2020-05-16

指数平滑模型与ARIMA模型在湖南省血吸虫病流行趋势预测中的应用

周杰,任光辉,贺宏斌,侯循亚,邓维成*   

  1. 湖南省血吸虫病防治所、WHO湖区血吸虫病防治合作中心、 血吸虫病免疫与传播控制湖南省重点实验室、国家临床重点专科(岳阳 414000)
  • 作者简介:周杰,男,主任医师。研究方向:血吸虫病防治
  • 基金资助:
    国家社科重大专项(16DA237);湖南省科技计划项目(2018JJ6020);湖南省卫生健康委科研计划项目(C2019061);湖南省岳阳市2015年巴陵人才工程科技创新创业人才团队项目(岳人才发[2015] 2号)

Abstract: Objective To predict the changes in the prevalence of Schistosoma japonicum infections in humans and livestock in Hunan Province using the exponential smoothing model and the ARIMA model. Methods The data pertaining to S. japonicum infections in humans and livestock in Hunan Province from 1957 to 2015 were collected, and the exponential smoothing model and the ARIMA model were created using the software Eviews and PASW Statistics 18.0. In addition, the effectiveness of these two models for the prediction of S. japonicum infections in humans and livestock in Hunan Province from 2016 to 2018 was evaluated. Results The exponential smoothing model and the ARIMA model had a high goodness of fit for prediction of S. japonicum infections in humans and livestock in Hunan Province from 1957 to 2015. There was a linear trend in the prevalence of S. japonicum infections in humans and livestock in Hunan Province from 1957 to 2015. The prevalence of S. japonicum infections in humans predicted with the Brown’s linear trend and the prevalence of S. japonicum infections in livestock predicted with the Holt’s linear trend in Hunan Province from 2016 to 2018 fitted better the actual data than the ARIMA model; however, prediction of the ARIMA model indicated that the endemic situation of schistosomiasis remained at a low level in Hunan Province. Conclusion At a low epidemic level, development of highly sensitive tools for monitoring schistosomiasis is urgently needed in Hunan Province to fit the current endemic situation, and the schistosomiasis control measures should be intensified to consolidate the control achievements.

Key words: Schistosomiasis, Exponential smoothing model, ARIMA model, Prediction, Hunan Province

摘要: 目的 应用指数平滑模型与ARIMA模型预测湖南省人畜血吸虫感染率变化趋势。方法 根据1957–2015年湖南省血吸虫病防治工作统计资料中的人畜感染率数据,建立指数平滑模型及ARIMA模型,并对2016–2018年湖南省人畜血吸虫感染率进行预测与效果评价。结果 1957–2015年湖南省人畜血吸虫感染率指数平滑模型与ARIMA模型均具有较好拟合效果。1957–2015年湖南省人畜血吸虫感染率具有线性趋势,2016–2018年人群感染率Brown’s线性趋势指数平滑模型与家畜感染率Holt’s线性趋势指数平滑模型预测值较2016–2018年人畜感染率ARIMA模型预测值更匹配于实际数据,但ARIMA模型提示湖南省血吸虫病可能仍处于低流行状态。结论 在血吸虫病低流行情况下,湖南省亟须研发高敏感度疾病监测技术,以匹配当前流行态势,并强化防控措施以巩固取得的防控成果。

关键词: 血吸虫病, 指数平滑模型, ARIMA模型, 预测, 湖南省

CLC Number: