[Application of the exponential smoothing model and ARIMA model in prediction of the endemic situation of schistosomiasis in Hunan Province].

Objective: To predict the changes in the prevalence of infections in humans and livestock in Hunan Province using the exponential smoothing model and the ARIMA model.

Methods: The data pertaining to infections in humans and livestock in Hunan Province from 1957 to 2015 were collected, and the exponential smoothing model and the ARIMA model were created using the software Eviews and PASW Statistics 18.0. In addition, the effectiveness of these two models for the prediction of infections in humans and livestock in Hunan Province from 2016 to 2018 was evaluated.

Results: The exponential smoothing model and the ARIMA model had a high goodness of fit for prediction of infections in humans and livestock in Hunan Province from 1957 to 2015. There was a linear trend in the prevalence of infections in humans and livestock in Hunan Province from 1957 to 2015. The prevalence of infections in humans predicted with the Brown's linear trend and the prevalence of infections in livestock predicted with the Holt's linear trend in Hunan Province from 2016 to 2018 fitted better the actual data than the ARIMA model; however, prediction of the ARIMA model indicated that the endemic situation of schistosomiasis remained at a low level in Hunan Province.

Conclusions: At a low epidemic level, development of highly sensitive tools for monitoring schistosomiasis is urgently needed in Hunan Province to fit the current endemic situation, and the schistosomiasis control measures should be intensified to consolidate the control achievements.