Comparison of three time series models in predicting the incidence of healthcare-associated infection / 中国感染控制杂志

Objective To compare and evaluate the effect of different time series models in predicting incidence of healthcare-associated infection (HAI), and explore the best model for predicting incidence of HAI.Methods Seasonal autoregressive integrated moving average (ARIMA) model, nonlinear autoregressive neural network (NARNN), and ARIMA-back propagation neural network (ARIMA-BPNN) combination model were constructed based on fitting dataset of monthly HAI incidence from 2011 to 2016 (72 months) in a tertiary first-class hospital in Shanghai, predicting dataset of monthly infection incidence from January to December 2017 were used to test the predictive effect of model, the predictive effect of different models was evaluated and compared.Results For the fitting dataset, mean absolute percentage error (MAPE) of ARIMA, NARNN, and ARIMA-BPNN combination model were 13.00％, 14.61％, and 11.95％respectively;and for the predicting dataset, MAPE of ARIMA, NARNN, and ARIMA-BPNN combination model were 15.42％, 26.31％, and 14.87％ respectively.Conclusion Three time series models can effectively predict the incidence of HAI, of which the ARIMA-BPNN combination model showed the best performance in fitting and predicting the occurrence of HAI in this hospital, and can provide data support for the hospital decision-making.

Application of nonlinear autoregressive neural network in predicting incidence tendency of hemorrhagic fever with renal syndrome / 中华流行病学杂志

Objective To explore the prospect of nonlinear autoregressive neural network in fitting and predicting the incidence tendency of hemorrhagic fever with renal syndrome (HFRS),in the mainland of China.Methods Monthly reported case series of HFRS in China from 2004 to 2013 were used to build both ARIMA and NAR neural network models,in order to predict the monthly incidence of HFRS in China in 2014.Fitness and prediction on the effects of these two models were compared.Results For the Fitting dataset,MAE,RMSE and MAPE of the ARIMA model were 148.058,272.077 and 12.678％ respectively,while the MAE,RMSE and MAPE of NAR neural network appeared as 119.436,186.671 and 11.778％ respectively.For the Predicting dataset,MAE,RMSE and MAPE of the ARIMA model appeared as 189.088,221.133 and 21.296％,while the MAE,RMSE and MAPE of the NAR neural network as 119.733,151.329 and 11.431％ respectively.Conclusion The NAR neural network showed better effects in fitting and predicting the incidence tendency of HFRS than using the traditional ARIMA model,in China.NAR neural network seemed to have strong application value in the prevention and control of HFRS.