Forecasting COVID-19 Case Trends Using SARIMA Models during the Third Wave of COVID-19 in Malaysia

Tan, Cia Vei; Singh, Sarbhan; Lai, Chee Herng; Zamri, Ahmed Syahmi Syafiq Md; Dass, Sarat C.; Aris, Tahir; Ibrahim, Hishamshah Mohd; Gill, Balvinder Singh

doi:10.3390/ijerph19031504

Cited by 26 publications

(14 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since, as stated above, such a circaseptan seasonality has not been observed previously in prior epidemics, this study cannot be compared to prior studies but similar SARIMA models have been recently tested using circaseptan seasonality to forecast 14 or 28 days [ 13 , 14 , 26 ] with comparable results.…”

Section: Resultsmentioning

confidence: 98%

“…Similar SARIMA models can be used for short-term predictions also in other regions and/or countries [13,14,26] but the best model, optimizing Bayesian Information Criterion (BIC) or Akaike Information Criterion (AIC), should be chosen based on available data. A similar model is currently used on the website epidata.it to forecast the next 14 days of new daily COVID-19 cases in Italy https://www.epidata.it/Italia/ARIMA.html, accessed on 28 May 2022.…”

Section: Discussionmentioning

confidence: 99%

“…Moreover, it has become increasingly important to define and refine epidemiological forecasting methods in order to adequately monitor population infection cases and deaths. SARIMA(X) models are an example of methods that have already been used for prediction in epidemiology, such as malaria [7], influenza-like illness [8], dengue hemorrhagic fever [9], West Nile virus [10], scarlet fever [11], human brucellosis [12], and recently, COVID-19 [13,14]. This paper presents the design and implementation of an approach based on autoregressive models to reliably forecast the spread of COVID-19 in Italian regions.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predicting the Spread of SARS-CoV-2 in Italian Regions: The Calabria Case Study, February 2020–March 2022

et al. 2022

View full text Add to dashboard Cite

Despite the stunning speed with which highly effective and safe vaccines have been developed, the emergence of new variants of SARS-CoV-2 causes high rates of (re)infection, a major impact on health care services, and a slowdown to the socio-economic system. For COVID-19, accurate and timely forecasts are therefore essential to provide the opportunity to rapidly identify risk areas affected by the pandemic, reallocate the use of health resources, design countermeasures, and increase public awareness. This paper presents the design and implementation of an approach based on autoregressive models to reliably forecast the spread of COVID-19 in Italian regions. Starting from the database of the Italian Civil Protection Department (DPC), the experimental evaluation was performed on real-world data collected from February 2020 to March 2022, focusing on Calabria, a region of Southern Italy. This evaluation shows that the proposed approach achieves a good predictive power for out-of-sample predictions within one week (R-squared > 0.9 at 1 day, R-squared > 0.7 at 7 days), although it decreases with increasing forecasted days (R-squared > 0.5 at 14 days).

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting the Spread of SARS-CoV-2 in Italian Regions: The Calabria Case Study, February 2020–March 2022

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Some covariates were considered to improve the model performances, such as state or district level daily case data, or the daily confirmed cases in a neighboring country. The results show that using smoothed data and independent covariates significantly improved the model accuracy [ 19 , 20 ].…”

Section: Prior Workmentioning

confidence: 99%

SARIMA Model Forecasting Performance of the COVID-19 Daily Statistics in Thailand during the Omicron Variant Epidemic

et al. 2022

View full text Add to dashboard Cite

This study aims to identify and evaluate a robust and replicable public health predictive model that can be applied to the COVID-19 time-series dataset, and to compare the model performance after performing the 7-day, 14-day, and 28-day forecast interval. The seasonal autoregressive integrated moving average (SARIMA) model was developed and validated using a Thailand COVID-19 open dataset from 1 December 2021 to 30 April 2022, during the Omicron variant outbreak. The SARIMA model with a non-statistically significant p-value of the Ljung–Box test, the lowest AIC, and the lowest RMSE was selected from the top five candidates for model validation. The selected models were validated using the 7-day, 14-day, and 28-day forward-chaining cross validation method. The model performance matrix for each forecast interval was evaluated and compared. The case fatality rate and mortality rate of the COVID-19 Omicron variant were estimated from the best performance model. The study points out the importance of different time interval forecasting that affects the model performance.

show abstract

“…The curve fitting distinguishes the traditional time series models [18,112,113] from the artificial intelligence (AI) time series models [112]. Traditional, statistical estimation methods include the maximum likelihood method, the conditional sum of least-squares and the ordinary sum of least-squares [108,109,114,115]. AI is an umbrella term for a collection of methods that searches a space for P r e -p r i n t an adequate solution.…”

Section: Introductionmentioning

confidence: 99%

Basic reproduction number projection for novel pandemics and variants. Ancestral SARS-CoV2 R0 projection

Benjamin¹

2022

Preprint

View full text Add to dashboard Cite

The recently derived Hybrid-Incidence Susceptible-Transmissible-Removed (HI-STR) prototype is a deterministic epidemic compartment model and an alternative to the Susceptible-Infected-Removed (SIR) model prototype. The HI-STR predicts that pathogen transmission depends on host population characteristics including population size, population density and some common host behavioural characteristics. The HI-STR prototype is applied to the ancestral Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV2) to show that the original estimates of the Coronavirus Disease 2019 (COVID-19) basic reproduction number ($\mathcal{R}_0$) for the United Kingdom (UK) could have been projected on the individual states of the United States of America (USA) prior to being detected in the USA. The Imperial College London (ICL) R0 estimate for the UK is projected onto each USA state. The difference between these projections and ICL estimates for USA states is either not statistically significant on the paired student t-test or epidemiologically insignificant. Projection provides a baseline for evaluating the real-time impact of an intervention. Sensitivity analysis was conducted because of considerable variance in parameter estimates across studies. Although the HI-STR predicts that increasing symptomatic ratio and inherently immune ratio reduce R0, relative to the uncertainty in the estimates of R0 for the ancestral SARS-CoV2, the projection is insensitive to the inherently immune ratio and the symptomatic ratio.

show abstract

Forecasting COVID-19 Case Trends Using SARIMA Models during the Third Wave of COVID-19 in Malaysia

Cited by 26 publications

References 27 publications

Predicting the Spread of SARS-CoV-2 in Italian Regions: The Calabria Case Study, February 2020–March 2022

Predicting the Spread of SARS-CoV-2 in Italian Regions: The Calabria Case Study, February 2020–March 2022

SARIMA Model Forecasting Performance of the COVID-19 Daily Statistics in Thailand during the Omicron Variant Epidemic

Basic reproduction number projection for novel pandemics and variants. Ancestral SARS-CoV2 R0 projection

Contact Info

Product

Resources

About