新型冠状病毒疫情分析及预测研究

Analysis and Prediction of NovelCoronavirus Epidemic

本文围绕着新型冠状病毒疫情进行了探究。首先收集了133个国家的时间序列数据,我们发现有不同国家表现出同一特点,为使国家更具代表性,我们从中删除了一些国家,最终确定了31个国家,根据其疫情发展特点,按照系统聚类的方法分类。其次,确定了疫情发展及管控的影响因素,采用灰色关联度分析和模糊综合评价的方法,对主要国家疫情管控效果进行了综合评价。最后,考虑到新型冠状病毒的潜伏期较长,建立了SEIR传染病模型,运用动力学方法模拟疫情发展过程,该过程的关键是寻找最优参数,我们采用最小二乘法。接着预测疫情发展情况并对其进行了F检验,得到F值大于理论值,表明模型拟合效果较好。

This article explores the novel coronavirus epidemic. Firstly, we collected time series data from 133 countries and found that there are different countries showing the same characteristics. In order to make the country more representative, we deleted some countries and finally identified 31 countries. According to the epidemic development characteristics, they were classified according to the method of systematic clustering. Secondly, the factors affecting the development and control of the epidemic were determined, and the methods of grey correlation analysis and fuzzy comprehensive evaluation were used to evaluate the effectiveness of epidemic control in major countries. Finally, considering the long incubation period of the new coronavirus, we established the SEIR model and the dynamic method was used to simulate the development process of the epidemic. The key to this process is to find the optimal parameters;we use the least squares method. Then the development of the epidemic was predicted and F test was performed on the prediction model;the F value was greater than the theoretical value, which indicated that the model fitting effect was good.