基于ES-GARCH方法的COVID-19疫情预测

Prediction of COVID-19 Epidemic Situation Based on ES-GARCH Method

摘要: 新型冠状病毒COVID-19导致的呼吸系统疾病在全球范围内爆发。为了更好地实施防控疫情的相关决策，研究人员利用统计模型预测疫情发展趋势以及评价这次传染病所带来的影响。在本文中，提出自回归条件异方差(ES-GARCH)的误差修正模型首次用来预测COVID-19的死亡人数和确诊病例，并综合了其他六种模型来探索预测精度最高的模型。首先，选取湖北省新冠肺炎的累计确诊人数和累计死亡人数，共两组数据集；其次，将数据集分别划分为训练集和测试集来建立模型；最后，使用评价指标对所使用模型的预测精度进行评估，结果显示本文中所提出的ES-GARCH模型的预测结果最佳。此外，为了确保模型的可靠性，选用意大利的两个数据集对模型进行验证，结果表现为该模型的预测性能优于其他六个预测模型。在文章的最后，使用ES-GARCH模型预测湖北省未来七天内COVID-19的累计确诊病例和累计死亡人数。通过验证，该模型适用于时间序列的短期预测，对预测全球疫情发展趋势有重要意义。

Abstract: Respiratory diseases caused by New Coronavirus COVID-19 break out all over the world. In order to better implement the relevant decision-making of epidemic prevention and control, researchers use statistical models to predict the trend of epidemic development and evaluate the impact of the epidemic. In this paper, an error correction model of autoregressive conditional heteroscedasticity (ES-GARCH) is proposed to predict the number of deaths and confirmed cases of COVID-19 for the first time, and other six models are integrated to explore the model with the highest prediction accuracy. First, the cumulative number of people diagnosed with COVID-19 and the cumulative number of deaths in Hubei Province were selected, and there were two data sets; secondly, we divide the data set into training set and test set to build the model; finally, the evaluation index is used to evaluate the prediction accuracy of the model, and the results show that the ES-GARCH model proposed in this paper has the best prediction results. In addition, in order to ensure the reliability of the model, two Italian data sets are selected to verify the model. The results show that the prediction performance of the model is better than the other six prediction models. At the end of the paper, ES-GARCH model is used to predict the cumulative number of confirmed cases and deaths of COVID-19 in Hubei Province in the next seven days. Through validation, the model is suitable for short-term prediction of time series, and is of great significance to predict the development trend of global epidemic.