基于ARIMA模型的新型冠状病毒肺炎疫情预测分析
作者：李凤英
来源：《价值工程》2020年第25期
![](https://wkrtcs.bdimg.com/rtcs/image?w=1027.00&md5sum=b679f24c6188a252a42828a128d6bd6a&sign=dc614fd004&rtcs_flag=1&rtcs_ver=4&l=webapp&bucketNum=614&ipr={"t":"img","r":"r_2","w":"1027.00","h":"1437.00","dataType":"jpeg","c":"word/media/image1.jpeg","imgOriW":1926,"imgOriH":2694})
![](https://wkrtcs.bdimg.com/rtcs/image?w=640.00&md5sum=b679f24c6188a252a42828a128d6bd6a&sign=dc614fd004&rtcs_flag=1&rtcs_ver=4&l=webapp&bucketNum=614&ipr={"t":"img","r":"r_2","w":"640.00","h":"1443.00","dataType":"jpeg","c":"word/media/image2.jpeg","imgOriW":1200,"imgOriH":2706})
![](https://wkrtcs.bdimg.com/rtcs/image?w=1027.00&md5sum=b679f24c6188a252a42828a128d6bd6a&sign=dc614fd004&rtcs_flag=1&rtcs_ver=4&l=webapp&bucketNum=614&ipr={"t":"img","r":"r_2","w":"1027.00","h":"842.00","dataType":"jpeg","c":"word/media/image3.jpeg","imgOriW":1926,"imgOriH":1578})
摘要：以湖北省2020/1/22-2020/3/24新型冠状病毒肺炎确诊数据为样本，利用 R语言构建求和自回归移动平均预测模型ARIMA（1，1，1），以1/22日-3/17日为训练数据， 3/18-3/24日为预测数据。运行模型发现，预测数据与真实数据拟合度高，检验效果显著。初期确诊病例数迅速上升，2月15日左右趋于稳定进入平稳缓慢期。将预测模型用于北京市，拟合效果良好，效果同样显著。充分说明ARIMA（1，1，1）模型稳健性良好，可用于新冠肺炎预测。
Abstract： Taking the confirmed data of COVID-19 in Hubei Province from 2020/1/22-2020/3/24 as samples， the summed autoregressive moving average prediction model ARIMA （1，1，1） was constructed by R language， 1/22 -3/17 Day is training data， 3/18-3/24 is prediction data. Running the model found that the prediction data and the real data have a high degree of fit， and the test effect is significant. 2/15 is about to stabilize and enter a stable and slow period. The model is used to predict the number of diagnosed people in Beijing. The fitting effect is good and the effect is also significant. It fully shows that the ARIMA （1，1，1） model has good robustness and can be used for COVID-19 prediction.
本文来源：https://www.2haoxitong.net/k/doc/a56cbfb53369a45177232f60ddccda38376be183.html
文档为doc格式