基于Logistic增长模型和相关性分析的疫情预测——以英国为例
Epidemic Prediction Based on Logistic Growth Model and Correlation Analysis—Taking the UK as an Example

目前，新冠肺炎病毒已经在世界各国传播。随着环境的变化和感染者人数的上升，病毒开始不断发生突变和进化，至今已经产生了11种变异毒株。由于发现不及时，各国在进行基因测序统计时，变种病毒往往已经开始传播了一段时间，缺乏初期传播的感染人数数据。本文以英国为例，根据病毒传播的特征，我们选择构造Logistic增长模型来模拟病毒感染过程，求得了各病毒的感染总人数曲线表达式，计算出了传播开始时间以及持续时间。此外，通过计算Pearson相关系数，对各类气候因素与病毒增长人数进行相关性分析，发现每日增长感染人数与温度和湿度相关性最大，且与温度呈负相关、湿度呈正相关。
Nowadays, the COVID-19 virus has spread all around the world. As the environment changed and the number of infected people rose, the virus began to mutate and evolve. So far, 11 mutated strains have emerged. Because of the delay in detection, the mutated strains have already started to spread for some time, giving rise to the lack of data on initial stage. Without losing generality, we shall discuss the case in the UK in this paper. According to the characteristics of the virus transmis-sion, we build Logistic growth models to simulate the process of the spread of virus and obtain the expression of the number of being infected in pace with time. As a result, we obtain the start time and duration of the virus. Additionally, we analyze the correlation between daily increase number and climatic factors in terms of the Pearson correlation coefficient and we find that the daily in-crease number is most correlated with temperature and humidity, specifically negative affected by temperature and positive affected by humidity.