Time series forecasting of the COVID-19 pandemic: a critical assessment in retrospect

Abstract

The COVID-19 pandemic is perceived by many to have run its course, and forecasting its progress is no longer a topic of much interest to policymakers and researchers as it once was. Nevertheless, in order to take lessons from this extraordinary two and a half years, it still makes sense to have a critical look at the vast body of literature formed thereon, and perform comprehensive analyses in retrospect. The present study is directed towards that goal. It is distinguished from others by encompassing all of the following features simultaneously: (i) time series of 10 of the most affected countries are considered; (ii) forecasting for two types of periods, namely days and weeks, are analyzed; (iii) a wide range of exponential smoothing, autoregressive integrated moving average, and neural network autoregression models are compared by means of automatic selection procedures; (iv) basic methods for benchmarking purposes as well as mathematical transformations for data adjustment are taken into account; and (v) several test and training data sizes are examined. Our experiments show that the performance of common time series forecasting methods is highly sensitive to parameter selection, bound to deteriorate dramatically as the forecasting horizon extends, and sometimes fails to be better than that of even the simplest alternatives. We contend that the reliableness of time series forecasting of COVID-19, even for a few weeks ahead, is open to debate. Policymakers must exercise extreme caution before they make their decisions utilizing a time series forecast of such pandemics.

Keywords

• time series forecasting
• coronavirus
• exponential smoothing
• autoregressive integrated moving average
• neural network autoregression

References

1. Abbasimehr, H., Paki, R., & Bahrini, A. (2022). A novel approach based on combining deep learning models with statistical methods for COVID-19 time series forecasting. Neural Computing and Applications, 34, 3135–3149. doi:10.1007/s00521-021-06548-9
2. Ahmad, G., Ahmed, F., Rizwan, M. S., Muhammad, J., Fatima, S. H., Ikram, A., & Zeeb, H. (2021). Evaluating data-driven methods for short-term forecasts of cumulative SARS-CoV2 cases. PLoS ONE, 16. doi:10.1371/journal.pone.0252147
3. Anadolu Agency. (2022). Many countries scrapping COVID-19 restrictions, thanks to high vaccination rates, low case incidence. Many countries scrapping COVID-19 restrictions, thanks to high vaccination rates, low case incidence. https://www.aa.com.tr/en/latest-on-coronavirus-outbreak/many-countries-scrapping-covid-19-restrictions-thanks-to-high-vaccination-rates-low-case-incidence/2500190 adresinden alındı
4. ArunKumar, K. E., Kalaga, D. V., Sai Kumar, C. M., Chilkoor, G., Kawaji, M., & Brenza, T. M. (2021). Forecasting the dynamics of cumulative COVID-19 cases (confirmed, recovered and deaths) for top-16 countries using statistical machine learning models: Auto-Regressive Integrated Moving Average (ARIMA) and Seasonal Auto-Regressive Integrated Moving Averag. Applied Soft Computing, 103. doi:10.1016/j.asoc.2021.107161
5. Aslan, I. H., Demir, M., Wise, M. M., & Lenhart, S. (2022). Modeling COVID-19: Forecasting and analyzing the dynamics of the outbreaks in Hubei and Turkey. Mathematical Methods in the Applied Sciences, 45, 6481–6494. doi:10.1002/mma.8181
6. Atchade, M. N., & Sokadjo, Y. M. (2022). Overview and cross-validation of COVID-19 forecasting univariate models. Alexandria Engineering Journal, 61, 3021–3036. doi:10.1016/j.aej.2021.08.028
7. Ballı, S. (2021). Data analysis of Covid-19 pandemic and short-term cumulative case forecasting using machine learning time series methods. Chaos, Solitons and Fractals, 142. doi:10.1016/j.chaos.2020.110512
8. Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time Series Analysis: Forecasting and Control.

9. Castillo, O., & Melin, P. (2020). Forecasting of COVID-19 time series for countries in the world based on a hybrid approach combining the fractal dimension and fuzzy logic. Chaos, Solitons and Fractals, 140. doi:10.1016/j.chaos.2020.110242
10. Ceylan, Z. (2020). Estimation of COVID-19 prevalence in Italy, Spain, and France. Science of the Total Environment, 729. doi:10.1016/j.scitotenv.2020.138817
11. Chimmula, V. K., & Zhang, L. (2020). Time series forecasting of COVID-19 transmission in Canada using LSTM networks. Chaos, Solitons and Fractals, 135. doi:10.1016/j.chaos.2020.109864
12. Coronavirus in the UK. (2022). Cases in United Kingdom. Cases in United Kingdom. https://coronavirus.data.gov.uk/details/cases?areaType=overview&areaName=United Kingdom adresinden alındı
13. Coroneo, L., Iacone, F., Paccagnini, A., & Santos Monteiro, P. (2022). Testing the predictive accuracy of COVID-19 forecasts. International Journal of Forecasting. doi:10.1016/j.ijforecast.2022.01.005
14. Daniyal, M., Tawiah, K., Muhammadullah, S., & Opoku-Ameyaw, K. (2022). Comparison of Conventional Modeling Techniques with the Neural Network Autoregressive Model (NNAR): Application to COVID-19 Data. Journal of Healthcare Engineering. doi:10.1155/2022/4802743
15. Devaraj, J., Madurai Elavarasan, R., Pugazhendhi, R., Shafiullah, G. M., Ganesan, S., Jeysree, A. K., . . . Hossain, E. (2021). Forecasting of COVID-19 cases using deep learning models: Is it reliable and practically significant? Results in Physics, 21. doi:10.1016/j.rinp.2021.103817
16. Doornik, J. A., Castle, J. L., & Hendry, D. F. (2022). Short-term forecasting of the coronavirus pandemic. International Journal of Forecasting, 38, 453–466. doi:10.1016/j.ijforecast.2020.09.003
17. Drews, M., Kumar, P., Singh, R. K., De La Sen, M., Singh, S. S., Pandey, A. K., . . . Srivastava, P. K. (2022). Model-based ensembles: Lessons learned from retrospective analysis of COVID-19 infection forecasts across 10 countries. Science of the Total Environment, 806. doi:10.1016/j.scitotenv.2021.150639
18. Eroğlu, Y. (2020). Forecasting Models for Covid-19 Cases of Turkey Using Artificial Neural Networks and Deep Learning. Journal of Industrial Engineering, 31, 354–372.
19. Fernandes, F., Stefenon, S. F., Seman, L. O., Nied, A., Ferreira, F. C., Subtil, M. C., . . . Leithardt, V. R. (2022). Long short-term memory stacking model to predict the number of cases and deaths caused by COVID-19. Journal of Intelligent and Fuzzy Systems, 42, 6221–6234. doi:10.3233/JIFS-212788
20. Guleryuz, D. (2021). Forecasting outbreak of COVID-19 in Turkey; Comparison of Box–Jenkins, Brown's exponential smoothing and long short-term memory models. Process Safety and Environmental Protection, 149, 927–935. doi:10.1016/j.psep.2021.03.032
21. Hewamalage, H., Bergmeir, C., & Bandara, K. (2021). Recurrent Neural Networks for Time Series Forecasting: Current status and future directions. International Journal of Forecasting, 37, 388–427. doi:10.1016/j.ijforecast.2020.06.008
22. Hyndman, R. J. (2021). Package `fpp3'. Package `fpp3'. https://cran.r-project.org/web/packages/fpp3/index.html adresinden alındı
23. Hyndman, R. J., & Athanasopoulos, G. (2021). Forecasting: Principles and Practice.
24. Hyndman, R. J., & Khandakar, Y. (2008). Automatic Time Series Forecasting: The forecast Package for R. Journal of Statistical Software, 27, 22. doi:10.18637/jss.v027.i03
25. Hyndman, R. J., Koehler, A. B., Ord, J. K., & Snyder, R. D. (2008). Forecasting with exponential smoothing: The state space approach.
26. Kırbaş, İ., Sözen, A., Tuncer, A. D., & Kazancıoğlu, F. Ş. (2020). Comparative analysis and forecasting of COVID-19 cases in various European countries with ARIMA, NARNN and LSTM approaches. Chaos, Solitons and Fractals, 138. doi:10.1016/j.chaos.2020.110015
27. Luo, J. (2021). Forecasting COVID-19 pandemic: Unknown unknowns and predictive monitoring. Technological Forecasting & Social Change, 166. doi:10.1016/j.techfore.2021.120602
28. Markeviciute, J., Bernataviciene, J., Levuliene, R., Medvedev, V., Treigys, P., & Venskus, J. (2021). Attention-based and time series models for short-term forecasting of COVID-19 spread. Computers, Materials and Continua, 70, 695–714. doi:10.32604/cmc.2022.018735
29. Mohanraj, G., Mohanraj, V., Marimuthu, M., Sathiyamoorthi, V., Luhach, A. K., & Kumar, S. (2022). Epidemic Prediction using Machine Learning and Deep Learning Models on COVID-19 Data. Journal of Experimental and Theoretical Artificial Intelligence. doi:10.1080/0952813X.2022.2058618
30. Niazkar, M., Türkkan, G. E., Niazkar, H. R., & Türkkan, Y. A. (2020). Assessment of three mathematical prediction models for forecasting the covid-19 outbreak in Iran and Turkey. Computational and Mathematical Methods in Medicine. doi:10.1155/2020/7056285
31. Nikolopoulos, K., Punia, S., Schäfers, A., Tsinopoulos, C., & Vasilakis, C. (2021). Forecasting and planning during a pandemic: COVID-19 growth rates, supply chain disruptions, and governmental decisions. European Journal of Operational Research, 290, 99–115. doi:10.1016/j.ejor.2020.08.001
32. O'Hara-Wild, M., Hyndman, R. J., & Wang, E. (2022). Package `fable'. Package `fable'. https://cran.r-project.org/web/packages/fable/index.html adresinden alındı
33. Our World in Data. (2022). Data on COVID-19. Data on COVID-19. https://github.com/owid/covid-19-data/tree/master/public/data adresinden alındı
34. Petropoulos, F., Makridakis, S., & Stylianou, N. (2022). COVID-19: Forecasting confirmed cases and deaths with a simple time series model. International Journal of Forecasting, 38, 439–452. doi:10.1016/j.ijforecast.2020.11.010
35. Rahimi, I., Chen, F., & Gandomi, A. H. (2021). A review on COVID-19 forecasting models. Neural Computing and Applications, 8. doi:10.1007/s00521-020-05626-8
36. Reuters. (2020). Turkey announces asymptomatic coronavirus case numbers for first time since July. Turkey announces asymptomatic coronavirus case numbers for first time since July. https://www.reuters.com/article/us-health-coronavirus-turkey-cases-idUSKBN2852W3 adresinden alındı
37. Rittel, H. W., & Webber, M. M. (1973). Dilemmas in a General Theory of Planning. Policy Sciences, 4, 155–169. doi:10.1007/BF01405730
38. Samanta, S., Prakash, P. K., & Chilukuri, S. (2022). MLTF: Model less time-series forecasting. Information Sciences, 593, 364–384. doi:10.1016/j.ins.2022.02.007
39. Talkhi, N., Akhavan Fatemi, N., Ataei, Z., & Jabbari Nooghabi, M. (2021). Modeling and forecasting number of confirmed and death caused COVID-19 in IRAN: A comparison of time series forecasting methods. Biomedical Signal Processing and Control, 66. doi:10.1016/j.bspc.2021.102494
40. Tan, C. V., Singh, S., Lai, C. H., Zamri, A. S., Dass, S. C., Aris, T. B., . . . Gill, B. S. (2022). Forecasting COVID-19 Case Trends Using SARIMA Models during the Third Wave of COVID-19 in Malaysia. International Journal of Environmental Research and Public Health, 19, 1–12. doi:10.3390/ijerph19031504
41. Toğa, G., Atalay, B., & Toksari, M. D. (2021). COVID-19 prevalence forecasting using Autoregressive Integrated Moving Average (ARIMA) and Artificial Neural Networks (ANN): Case of Turkey. Journal of Infection and Public Health, 14, 811–816. doi:10.1016/j.jiph.2021.04.015
42. Turkish Ministry of Health. (2022). Genel Koronavirüs Tablosu (in Turkish). Genel Koronavirüs Tablosu (in Turkish). https://covid19.saglik.gov.tr/TR-66935/genel-koronavirus-tablosu.html adresinden alındı
43. Wang, Y., Yan, Z., Wang, D., Yang, M., Li, Z., Gong, X., . . . Wang, Y. (2022). Prediction and analysis of COVID-19 daily new cases and cumulative cases: times series forecasting and machine learning models. BMC Infectious Diseases, 22, 1–12. doi:10.1186/s12879-022-07472-6
44. World Health Organization. (2022). The COVID-19 pandemic is nowhere near over. The COVID-19 pandemic is nowhere near over. https://unric.org/en/the-covid-19-pandemic-is-nowhere-near-over-who adresinden alındı
45. World Health Organization. (2022). The end of the COVID-19 pandemic is in sight. The end of the COVID-19 pandemic is in sight. https://news.un.org/en/story/2022/09/1126621 adresinden alındı
46. Zeroual, A., Harrou, F., Dairi, A., & Sun, Y. (2020). Deep learning methods for forecasting COVID-19 time-series data: A comparative study. Chaos, Solitons and Fractals, 140. doi:10.1016/j.chaos.2020.110121
47. Zhang, Q., & Yi, G. Y. (2022). Sensitivity analysis of error-contaminated time series data under autoregressive models with the application of COVID-19 data. Journal of Applied Statistics. doi:10.1080/02664763.2022.2034760.