基于2014-2015年埃博拉疫情数据的统计建模与重要参数分析

发布时间：2018-08-08 20:12

【摘要】：世界卫生组织研究报告表明,传染病是人类的第一杀手,并不断地威胁着人类的生命健康。如2003年SARS的爆发,给人类和经济带来了巨大的危害和损失。由于传染病不能采取实验形式研究,因此,对传染病的传播途径、传播机制以及流行规律更需要理论分析和研究。传染病动力学就是针对传染病的流行发展规律进行定量研究的一种分析方法。本文根据2014-2015年西非埃博拉病毒数据,采用极大似然估计方法,运用三类传染病动力学模型(SIR模型、SEIR模型和SEIHFR模型),分别对总体、几内亚、塞拉利昂和利比里亚的疫情进行实证分析。通过R语言编程给出重要参数估计值,埃博拉病毒累计病例拟合曲线。经过比较,增加了住院期和埋葬期两个状态的SEIHFR模型,更符合埃博拉病毒的传播机理和流行规律。研究结论表明,控制措施干预之前,所有情形下的基本再生数0都大于1,说明总体、几内亚、塞拉利昂和利比里亚的埃博拉病毒始终保持进而形成地方病;控制措施干预之后,由于所有情形下的控制系数均大于0,则其有效再生数e()都随时间的增加而呈指数下降,并降低为1。SEIHFR模型在总体、几内亚、塞拉利昂和利比里亚疫情的有效再生数降低为1的时间分别在2014年12月25日、2014年11月19日、2014年11月29日以及2014年10月23日。表明政府的控制干预措施对总体、几内亚、塞拉利昂和利比里亚的埃博拉疫情的传播抑制是成功有效的。从拟合曲线中可以发现,SEIHFR模型更适用于总体、几内亚和利比里亚的疫情分析;对于塞拉利昂疫情,SIR模型的效果会更好。
[Abstract]:The World Health Organization (WHO) report shows that infectious diseases are the first killer of human beings and threaten human life and health. For example, the outbreak of SARS in 2003 brought great harm and loss to human and economy. Because infectious diseases can not be studied in experimental form, it is necessary to analyze and study the transmission path, transmission mechanism and epidemic law of infectious diseases. The dynamics of infectious diseases is an analytical method for the quantitative study of epidemic development law of infectious diseases. Based on the Ebola virus data from West Africa in 2014-2015, using the maximum likelihood estimation method, using three infectious disease dynamics models (SIR model, seir model and SEIHFR model), respectively, An empirical analysis of the outbreaks in Sierra Leone and Liberia was carried out. The estimation value of important parameters was given by programming in R language, and the fitting curve of Ebola virus accumulative cases was obtained. After comparison, the SEIHFR model of two states of hospitalization and burial period was added, which was more consistent with the spreading mechanism and epidemic rule of Ebola virus. The results showed that before the intervention of control measures, the number of basic regenerations in all cases was greater than 1, indicating that the Ebola virus in Guinea, Sierra Leone and Liberia remained endemic in general, and that after the intervention of the control measures, Since the control coefficient is greater than 0 in all cases, the effective reproduction number e () decreases exponentially with the increase of time, and is reduced to the 1.SEIHFR model in general, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, Guinea, The number of effective rebirths in Sierra Leone and Liberia decreased to 1 on 25 December 2014, 19 November 2014, 29 November 2014 and 23 October 2014, respectively. This suggests that government control interventions have been successful in curbing the spread of Ebola in Guinea, Sierra Leone and Liberia as a whole. It can be found from the fitting curve that the SEIHFR model is more suitable for the overall analysis of the epidemic situation in Guinea and Liberia, and the effect of the Sir model on the epidemic situation in Sierra Leone is better.
【学位授予单位】：暨南大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R512.8

【相似文献】