Application of a Novel Grey Self-Memory Coupling Model to Forecast the Incidence Rates of Two Notifiable Diseases in China: Dysentery and Gonorrhea

Abstract

ObjectiveIn this study, a novel grey self-memory coupling model was developed to forecast the incidence rates of two notifiable infectious diseases (dysentery and gonorrhea); the effectiveness and applicability of this model was assessed based on its ability to predict the epidemiological trend of infectious diseases in China.MethodsThe linear model, the conventional GM(1,1) model and the GM(1,1) model with self-memory principle (SMGM(1,1) model) were used to predict the incidence rates of the two notifiable infectious diseases based on statistical incidence data. Both simulation accuracy and prediction accuracy were assessed to compare the predictive performances of the three models. The best-fit model was applied to predict future incidence rates.ResultsSimulation results show that the SMGM(1,1) model can take full advantage of the systematic multi-time historical data and possesses superior predictive performance compared with the linear model and the conventional GM(1,1) model. By applying the novel SMGM(1,1) model, we obtained the possible incidence rates of the two representative notifiable infectious diseases in China.ConclusionThe disadvantages of the conventional grey prediction model, such as sensitivity to initial value, can be overcome by the self-memory principle. The novel grey self-memory coupling model can predict the incidence rates of infectious diseases more accurately than the conventional model, and may provide useful references for making decisions involving infectious disease prevention and control.

A novel grey model based on traditional Richards model and its application in COVID-19

“…Guo et al. [26] used traditional GM(1,1) and SMGM(1,1) based on self-memory principle to predict the incidence of three typical infectious diseases in China. Wang et al.…”

Early Warning of Infectious Diseases in Hospitals Based on Multi-Self-Regression Deep Neural Network

“…For example, Feng et al [ 11 ] obtained the latent data by estimation, which affects accuracy, thus restricting the scope of application. In some studies, the autoregressive integrated moving average (ARIMA) model [ 12 , 13 ], linear regression [ 14 ], moment estimation [ 15 ], hidden Markov model [ 16 ], and grey self-memory coupling model [ 17 ] were adopted. However, there is a limit to the scope of application for each of these mathematical models, which means they are often suited to a single or certain type of disease.…”

Research of Combined ES-BP Model in Predicting Syphilis Incidence 1982-2020 in Mainland China

“…For time-series models, Zhang et al ( 6 ) used the ARIMA and ARIMAX models to predict the incidence of syphilis in mainland China from 2005 to 2012. Guo et al ( 8 ) used the conventional GM ( 1 , 1 ) model and GM ( 1 , 1 ) model with the self-memory principle (SMGM ( 1 , 1 ) model) to forecast the incidence of syphilis in mainland China. Li et al ( 9 ) explored the feasibility of using the grey GM ( 1 , 1 ) model to predict the incidence of syphilis.…”