上海市流行性腮腺炎疫情时间序列模型建立的初探

吴琳琳; 孙晓冬; 胡家瑜; 李智; 杨建萍

doi:10.19428/j.cnki.sjpm.2018.18669

上海市流行性腮腺炎疫情时间序列模型建立的初探

Autoregressive integrated moving average model applied on prediction of mumps in Shanghai

摘要

摘要:
目的探讨时间序列模型在流行性腮腺炎（流腮）预测中的应用，建立上海市流腮发病的预测模型，预测2017年上海市流腮发病趋势。
方法收集中国疾病监测信息报告系统中的上海市2005年1月-2016年12月流腮月报告发病资料，使用SPSS软件进行建模，考虑季节因素建立ARIMA（Autoregressive Integrated Moving Average）乘积季节预测模型，并用所建模型预测上海市2017年流腮发病趋势。
结果ARIMA（1，0，0）（1，1，0）₁₂可较好地拟合流腮发病的时间序列趋势，对2005-2016年流腮发病数预测值与实际值吻合程度高，平均相对误差为8.79%，2017年流腮预测病例数为2 656例。
结论ARIMA乘积季节模型可较好地拟合流腮发病的时间序列趋势；与2016年相比，预测2017年流腮报告发病数相对平稳。

Abstract:
ObjectiveTo explore the application of time series analysis for mumps prediction, and to establish ARIMA model to predict mumps cases per month in Shanghai.
MethodsData were collected on monthly reports about mumps cases from China Information System for Disease Control and Prevention between January 2005 and December 2016;ARIMA model was established for predicting the trend of mumps in 2017 by SPSS.
ResultsARIMA (1, 0, 0)(1, 1, 0)₁₂ model could well fit the time series trend for mumps occurrence.The predicted values of monthly cases well matched the actual cases and the average relative error was 8.79%;2 656 cases were predicted in 2017.
ConclusionARIMA model could well simulate the changing trend of mumps cases in time series.Compared with the cases in 2016, mumps cases reported in number is predicted to be relatively stable in 2017.

HTML全文

参考文献(9)

施引文献

资源附件(0)