Time Prediction Models for Echinococcosis Based on Gray System Theory and Epidemic Dynamics

Abstract

Echinococcosis, which can seriously harm human health and animal husbandry production, has become an endemic in the Xinjiang Uygur Autonomous Region of China. In order to explore an effective human Echinococcosis forecasting model in Xinjiang, three grey models, namely, the traditional grey GM(1,1) model, the Grey-Periodic Extensional Combinatorial Model (PECGM(1,1)), and the Modified Grey Model using Fourier Series (FGM(1,1)), in addition to a multiplicative seasonal ARIMA(1,0,1)(1,1,0)4 model, are applied in this study for short-term predictions. The accuracy of the different grey models is also investigated. The simulation results show that the FGM(1,1) model has a higher performance ability, not only for model fitting, but also for forecasting. Furthermore, considering the stability and the modeling precision in the long run, a dynamic epidemic prediction model based on the transmission mechanism of Echinococcosis is also established for long-term predictions. Results demonstrate that the dynamic epidemic prediction model is capable of identifying the future tendency. The number of human Echinococcosis cases will increase steadily over the next 25 years, reaching a peak of about 1250 cases, before eventually witnessing a slow decline, until it finally ends.

Estimation of COVID-19 prevalence in Italy, Spain, and France

“…Recently, different statistical methods such as time series models (Kurbalija et al, 2014), multivariate linear regression (Thomson et al, 2006), grey forecasting models (Wang et al, 2018a;Zhang et al, 2017), backpropagation neural networks Ren et al, 2013;Zhang et al, 2013), and simulation models (Nsoesie et al, 2013;Orbann et al, 2017) were used to predict epidemic cases. Epidemics are affected by many different factors.…”

Time series analysis of cumulative incidences of typhoid and paratyphoid fevers in China using both Grey and SARIMA models

“…The GM assumes all data are positive (consistent with cumulative incidence) and sampling frequency of the time series is fixed (true for annual or monthly datasets). The grey model does not require errors in normal distribution and is hardly limited by small sample size [ 20 ]. Seasonal Autoregressive Integrated Moving Average (SARIMA) model, simultaneously taking general trend, periodic pattern, and random disturbance into consideration, has been widely applied in the research on various infectious diseases with periodic pattern, e.g.…”

Forecasting the Spreading of COVID-19 across Nine Countries from Europe, Asia, and the American Continents Using the ARIMA Models

“…Some examples are represented by statistical methods aiming to predict epidemic cases. These include time series [ 16 ], or simulation models [ 17 , 18 ], multivariate linear regression [ 19 ], backpropagation neural network [ 20 , 21 , 22 ], and gray forecasting [ 23 , 24 ].…”