A time series analysis on under 5 mortality rate in Zhejiang Province / 预防医学

ABSTRACT

Objective To study the temporal distribution regular pattern of under 5 mortality rate(U5MR)from 1 998 to 201 4 in Zhejiang Province,and to predict the under 5 mortality rate in 201 5.Methods A time series ARIMA (p,d,q) forecasting model for U5MR was conducted using IBM SPSS Statistics 20.0 statistical analysis software.Results The UMAR showed downward trend.The ARIMA(2,1 ,2)model of U5MR from 1 998 to 201 4 in Zhejiang Province is yt =-0.696 +0.636yt -1 +0.024yt -2 +0.340yt -3 +αt -0.003αt -1 +0.997αt -2 ,and the model fitting was good.Each of the actual mortality was consistent with the trend of model prediction,and was within the 95% confidence interval.The predicted value of U5MR was 4.08‰ (95% CI1 .52‰ -6.64‰)in 201 5.Conclusion Time series analysis is an effective way to analyze the temporal distribution regular pattern of U5MR,which could be used for short -term prediction.