Short-term prediction of COVID-19 spread using grey rolling model optimized by particle swarm optimization

Abstract

The prediction of the spread of coronavirus disease 2019 (COVID-19) is vital in taking preventive and control measures to reduce human health damage. The Grey Modeling (1,1) is a popular approach used to construct a predictive model with a small-sized data set. In this study, a hybrid model based on grey prediction and rolling mechanism optimized by particle swarm optimization algorithm (PSO) was applied to create short-term estimates of the total number of confirmed COVID-19 cases for three countries, Germany, Turkey, and the USA. A rolling mechanism that updates data in equal dimensions was applied to improve the forecasting accuracy of the models. The PSO algorithm was used to optimize the Grey Modeling parameters (1,1) to provide more robust and efficient solutions with minimum errors. To compare the accuracy of the predictive models, a nonlinear autoregressive neural network (NARNN) was also developed. According to the analysis results, Grey Rolling Modeling (1,1) optimized by PSO algorithm performs better than the classical Grey Modeling (1,1), Grey Rolling Modelling (1,1), and NARNN models for predicting the total number of confirmed COVID-19 cases. The present study can provide an important basis for countries to allocate health resources and formulate epidemic prevention policies effectively.

A novel grey model based on Susceptible Infected Recovered Model: A case study of COVD-19

“…Due to its excellent and stable prediction performance for small sample data, many scholars applied the grey prediction model to the prediction of COVID-19. For example, Zeynep [25] used a classical GM(1,1) model with the rolling mechanism to predict the number of confirmed cases in Germany, Turkey, and the United States. Liu et al.…”

Dual attention-based sequential auto-encoder for Covid-19 outbreak forecasting: A case study in Vietnam

“…To effectively deal with the non-linear patterns of daily reported infectious cases, there are recent attempts of Ceylan ( Ceylan, 2021 ) and Hansun et al ( Hansun et al, 2021 ) have demonstrated the effectiveness of integrating NARNN architecture. These NARRN-based models ( Ceylan, 2021 , Hansun et al, 2021 ) support to model the flexible time-dependent growth patterns of Covid-19 diseases in different real-world datasets. In general, the NARNN approach is basically considered as an initial artificial neural network based technique.…”

“…This type of infectious data is majorly influenced by different external aspects. In order to deal with the non-linear and complex sequential data pattern learning, there are several studies ( Ceylan, 2021 , Wu et al, 2015 , Yu et al, 2014 ) have applied the NARNN-based approach to enable the capability of predictive system in performing non-linear and time-series based data analysis problem. In this approach, the temporal information is efficiently preserved from the input time-dependent observations through the neural network based learning paradigm.…”

“…In general, the NARNN is considered as a neural network based approach which explicitly supports to obtain the temporal information from time series data through utilizing different sequences of a given dataset and train them with corresponding parameter weights through the back-propagation learning procedure. Recent enhancements ( Ceylan, 2021 , Hansun et al, 2021 ) in the integration of NARNN with predictive systems have demonstrated remarkable performances in dealing with short-term Covid-19 outbreak forecasting problem. The successes of NARNN in complex time series based prediction task has clearly indicated that deep neural network is the potential and key factor for achieving better performance in temporal/sequential data representation learning and prediction problem.…”

Epidemic Peaks Forecasting on Re-emerging Diseases in Elderly People using the Grey Disaster Model

“…are widely used in medicines [4][5][6][7][8][9], and other areas such as petroleum, geological engineering, industry, management, marketing, and agriculture [10][11][12][13][14][15][16][17][18][19][20][21][22][23]. In medicines and public health, the spread of diseases is affected by several uncertain factors, so the Grey theory with dynamic changes is suitable to be used.…”