Research Article
BibTex RIS Cite

Investigation of Covid 19 Cases in Sivas Province with Time Series Analysis

Year 2025, Volume: 47 Issue: 4, 32 - 39, 31.12.2025

Esra Akaydın Gültürk , Emek Güldoğan

Abstract

Objective: The objective of this study is to compare the prediction performance of classical statistical, machine learning, and deep learning-based modeling approaches by performing time series analysis of monthly case numbers in Sivas province during the COVID-19 pandemic. Thus, the aim is to determine the most suitable model that can more accurately predict the future course of the pandemic.
Method: The study used confirmed COVID-19 case data from March 1, 2020, to August 1, 2022. The data were obtained from the records of the Turkish Ministry of Health and the Sivas Provincial Health Directorate. The stationarity of the series was assessed using Augmented Dickey–Fuller (ADF) and KPSS tests, and a logarithmic transformation was applied to reduce variance imbalance. The data were divided into a 70% training set, a 10% validation set, and a 20% test set. Classical statistical SARIMAX, machine learning-based Prophet, and deep learning-based LSTM models were used in the analyses, and performance comparisons were made using RMSE and SMAPE metrics.
Findings: According to the results, the LSTM model showed the lowest error rate with RMSE=1.975 and SMAPE=109.9%. The log-transformed SARIMAX model (RMSE=2.093; SMAPE=114.3%) ranked second, while the Prophet model (RMSE=5.290; SMAPE=116.6%) ranked third. The error rates of models built with raw data were significantly higher.
Conclusion: While logarithmic transformation improved the performance of classical models, the LSTM model provided the highest accuracy for complex and non-linear time series such as COVID-19. This result demonstrates that deep learning approaches offer a powerful alternative for epidemiological forecasting and can significantly contribute to decision-making in local health planning and resource management.

Keywords

COVID-19 Time Series Analysis SARIMAX Prophet LSTM

References

  • 1. Abade A, Porto LF, Scholze AR, et al. A comparative analysis of classical and machine learning methods for forecasting TB/HIV co-infection. Scientific Reports. 2024;14(1):18991. doi:https://doi.org/10.1038/s41598-024-69580-4
  • 2. Çelik Ş, Yurtbay S, Tekin YK, Korkmaz İ, Çelik P. Covid-19 Pandemisinin Acil Servis Yüküne Etkisi. Van Tıp Dergisi. 2022;29(3):303-308. doi:https://doi.org/10.5505/vtd.2022.87405
  • 3. Rehman AU, Mian SH, Usmani YS, Abidi MH, Mohammed MK. Modeling consequences of COVID-19 and assessing its epidemiological parameters: a System Dynamics Approach. MDPI; 2023:260.
  • 4. Wang Y, Xu C, Yao S, et al. Estimating the COVID-19 prevalence and mortality using a novel data-driven hybrid model based on ensemble empirical mode decomposition. Scientific Reports. 2021;11(1):21413. doi:https://doi.org/10.1038/s41598-021-00948-6
  • 5. Özen NS, Saraç S, Koyuncu M. COVID-19 vakalarının makine öğrenmesi algoritmaları ile tahmini: Amerika Birleşik Devletleri örneği. Avrupa Bilim ve Teknoloji Dergisi. 2021;(22):134-139. doi: https://doi.org/10.31590/ejosat.855113
  • 6. Guleryuz D. 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. 2021;149:927-935. doi:https://doi.org/10.1016/j.psep.2021.03.032
  • 7. Zhao W, Sun Y, Li Y, Guan W. Prediction of COVID-19 data using hybrid modeling approaches. Frontiers in public health. 2022;10:923978. doi:https://doi.org/10.3389/fpubh.2022.923978
  • 8. Sevli O, Gülsoy VGB. Covid-19 salgınına yönelik zaman serisi verileri ile Prophet model kullanarak makine öğrenmesi temelli vaka tahminlemesi. Avrupa Bilim ve Teknoloji Dergisi. 2020;(19):827-835. doi:https://doi.org/10.31590/ejosat.766623
  • 9. Taylor SJ, Letham B. Forecasting at scale. The American Statistician. 2018;72(1):37-45. doi:https://doi.org/10.7287/peerj.preprints.3190v2
  • 10. Kaya U, Akba F, Medeni İ, Medeni T. Covid-19 öncesi ve sonrasındaki Bitcoin fiyat değişimlerinin Makine Öğrenmesi, zaman serileri analizi ve derin öğrenme yöntemleriyle değerlendirilmesi. Bilişim Teknolojileri Dergisi. 2020;13(3):341-355. doi:https://doi.org/10.17671/gazibtd.648424
  • 11. Napoli AM, Smith-Shain R, Lin T, Baird J. The accuracy of predictive analytics in forecasting emergency department volume before and after onset of COVID-19. Western Journal of Emergency Medicine. 2023;25(1):61. doi:https://doi.org/10.5811/westjem.61059
  • 12. Baccega D, Castagno P, Fernández Anta A, Sereno M. Enhancing COVID-19 forecasting precision through the integration of compartmental models, machine learning and variants. Scientific Reports. 2024;14(1):19220. doi:https://doi.org/10.1038/s41598-024-69660-5
  • 13. Farhat F, Sohail SS, Alam MT, et al. COVID-19 and beyond: leveraging artificial intelligence for enhanced outbreak control. Frontiers in Artificial Intelligence. 2023;6:1266560. doi:https://doi.org/10.3389/frai.2023.1266560
  • 14. ArunKumar K, Kalaga DV, Kumar CMS, Kawaji M, Brenza TM. Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends. Alexandria engineering journal. 2022;61(10):7585-7603. doi:https://doi.org/10.1016/j.aej.2022.01.011
  • 15. Deschepper M, Eeckloo K, Malfait S, Benoit D, Callens S, Vansteelandt S. Prediction of hospital bed capacity during the COVID− 19 pandemic. BMC health services research. 2021;21(1):468. doi:https://doi.org/10.1186/s12913-021-06492-3
  • 16. Ódor G, Karsai M. Epidemic-induced local awareness behavior inferred from surveys and genetic sequence data. Nature Communications. 2025;16(1):4758. doi:https://doi.org/10.48550/arxiv.2406.09983
  • 17. Ali AO, Palanichamy N, Haw SC, Gopal S. Performance Evaluation on COVID-19 Prediction using Machine Learning Models. Journal of Informatics and Web Engineering. 2025;4(2):64-76. doi:https://doi.org/10.33093/jiwe.2025.4.2.5
  • 18. K Abdul Hamid AA, Wan Mohamad Nawi WIA, Lola MS, et al. Improvement of time forecasting models using machine learning for future pandemic applications based on COVID-19 data 2020–2022. Diagnostics. 2023;13(6):1121. doi:https://doi.org/10.3390/diagnostics13061121
  • 19. Zhao D, Zhang R, Zhang H, He S. Prediction of global omicron pandemic using ARIMA, MLR, and Prophet models. Scientific Reports. 2022;12(1):18138. doi: https://doi.org/10.1038/s41598-022-23154-4
  • 20. Nassiri H, Mohammadpour SI, Dahaghin M. How do the smart travel ban policy and intercity travel pattern affect COVID-19 trends? Lessons learned from Iran. Plos one. 2022;17(10):e0276276. doi:https://doi.org/10.1371/journal.pone.0276276
  • 21. Li X, Zhang X. A comparative study of statistical and machine learning models on carbon dioxide emissions prediction of China. Environmental Science and Pollution Research. 2023;30(55):117485-117502. doi:https://doi.org/10.1007/s11356-023-30428-5
  • 22. Wang Y, Xu C, Wu W, et al. Time series analysis of temporal trends in hemorrhagic fever with renal syndrome morbidity rate in China from 2005 to 2019. Scientific Reports. 2020;10(1):9609. doi:https://doi.org/10.1038/s41598-020-66758-4
  • 23. Chen Q, Zheng X, Shi H, et al. Prediction of influenza outbreaks in Fuzhou, China: comparative analysis of forecasting models. BMC Public Health. 2024;24(1):1399. doi:https://doi.org/10.1186/s12889-024-18583-x
  • 24. Zhou L, Zhao P, Wu D, Cheng C, Huang H. Time series model for forecasting the number of new admission inpatients. BMC medical informatics and decision making. 2018;18(1):39. doi:https://doi.org/10.1186/s12911-018-0616-8
  • 25. Jäger WS, Nagler T, Czado C, McCall RT. A statistical simulation method for joint time series of non-stationary hourly wave parameters. Coastal Engineering. 2019;146:14-31. doi:https://doi.org/10.1016/j.coastaleng.2018.11.003
  • 26. Sharma K, Bhalla R, Ganesan G. Time Series Forecasting Using FB-Prophet. 2022:59-65.
  • 27. Zunic E, Korjenic K, Hodzic K, Donko D. Application of facebook's prophet algorithm for successful sales forecasting based on real-world data. arXiv preprint arXiv:200507575. 2020;doi: https://doi.org/10.5121/ijcsit.2020.12203
  • 28. Vo NN, He X, Liu S, Xu G. Deep learning for decision making and the optimization of socially responsible investments and portfolio. Decision Support Systems. 2019;124:113097. doi:https://doi.org/10.1016/j.dss.2019.113097
  • 29. Woltmann L, Deepe J, Hartmann C, Lehner W. evalPM: a framework for evaluating machine learning models for particulate matter prediction. Environmental Monitoring and Assessment. 2023;195(12):1491. doi:https://doi.org/10.1007/s10661-023-11996-y
  • 30. Bandara K, Bergmeir C, Smyl S. Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach. Expert systems with applications. 2020;140:112896. doi:https://doi.org/10.1016/j.eswa.2019.112896

Sivas İlinde Covid 19 Vakalarının Zaman Serileri Analizi İle İncelenmesi

Year 2025, Volume: 47 Issue: 4, 32 - 39, 31.12.2025

Esra Akaydın Gültürk , Emek Güldoğan

Abstract

Amaç: Bu çalışmanın amacı, COVID-19 pandemisi süresince Sivas iline ait aylık vaka sayılarının zaman serisi analizini gerçekleştirerek, klasik istatistiksel, makine öğrenmesi ve derin öğrenme tabanlı modelleme yaklaşımlarının tahmin performanslarını karşılaştırmaktır. Böylece, salgının gelecekteki seyrini daha doğru öngörebilecek en uygun modelin belirlenmesi hedeflenmiştir.
Yöntem: Çalışmada 1 Mart 2020 – 1 Ağustos 2022 tarihleri arasındaki doğrulanmış COVID-19 vaka verileri kullanılmıştır. Veriler, T.C. Sağlık Bakanlığı ve Sivas İl Sağlık Müdürlüğü kayıtlarından elde edilmiştir. Serinin durağanlığı Augmented Dickey Fuller (ADF) ve KPSS testleriyle değerlendirilmiş, varyans dengesizliğini azaltmak amacıyla logaritmik dönüşüm uygulanmıştır. Veriler %70 eğitim, %10 doğrulama ve %20 test kümesi olarak ayrılmıştır. Analizlerde klasik istatistiksel SARIMAX, makine öğrenmesi tabanlı Prophet ve derin öğrenme temelli LSTM modelleri kullanılmış, performans karşılaştırmaları RMSE ve SMAPE ölçütleriyle yapılmıştır.
Bulgular: Sonuçlara göre, LSTM modeli RMSE=1.975 ve SMAPE=%109.9 ile en düşük hata oranını göstermiştir. Log-dönüştürülmüş SARIMAX modeli (RMSE=2.093; SMAPE=%114.3) ikinci sırada, Prophet modeli (RMSE=5.290; SMAPE=%116.6) ise üçüncü sırada yer almıştır. Ham verilerle kurulan modellerin hata oranları belirgin biçimde daha yüksek bulunmuştur.
Sonuç: Logaritmik dönüşüm, klasik modellerin performansını artırmakla birlikte, COVID-19 gibi karmaşık ve doğrusal olmayan zaman serilerinde en yüksek doğruluğu LSTM modeli sağlamıştır. Bu sonuç, derin öğrenme yaklaşımlarının epidemiyolojik öngörülerde güçlü bir alternatif sunduğunu ve yerel sağlık planlaması ile kaynak yönetiminde karar vericilere önemli katkı sağlayabileceğini göstermektedir.

Keywords

COVID-19 Zaman Serisi Analizi SARIMAX Prophet LSTM

References

  • 1. Abade A, Porto LF, Scholze AR, et al. A comparative analysis of classical and machine learning methods for forecasting TB/HIV co-infection. Scientific Reports. 2024;14(1):18991. doi:https://doi.org/10.1038/s41598-024-69580-4
  • 2. Çelik Ş, Yurtbay S, Tekin YK, Korkmaz İ, Çelik P. Covid-19 Pandemisinin Acil Servis Yüküne Etkisi. Van Tıp Dergisi. 2022;29(3):303-308. doi:https://doi.org/10.5505/vtd.2022.87405
  • 3. Rehman AU, Mian SH, Usmani YS, Abidi MH, Mohammed MK. Modeling consequences of COVID-19 and assessing its epidemiological parameters: a System Dynamics Approach. MDPI; 2023:260.
  • 4. Wang Y, Xu C, Yao S, et al. Estimating the COVID-19 prevalence and mortality using a novel data-driven hybrid model based on ensemble empirical mode decomposition. Scientific Reports. 2021;11(1):21413. doi:https://doi.org/10.1038/s41598-021-00948-6
  • 5. Özen NS, Saraç S, Koyuncu M. COVID-19 vakalarının makine öğrenmesi algoritmaları ile tahmini: Amerika Birleşik Devletleri örneği. Avrupa Bilim ve Teknoloji Dergisi. 2021;(22):134-139. doi: https://doi.org/10.31590/ejosat.855113
  • 6. Guleryuz D. 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. 2021;149:927-935. doi:https://doi.org/10.1016/j.psep.2021.03.032
  • 7. Zhao W, Sun Y, Li Y, Guan W. Prediction of COVID-19 data using hybrid modeling approaches. Frontiers in public health. 2022;10:923978. doi:https://doi.org/10.3389/fpubh.2022.923978
  • 8. Sevli O, Gülsoy VGB. Covid-19 salgınına yönelik zaman serisi verileri ile Prophet model kullanarak makine öğrenmesi temelli vaka tahminlemesi. Avrupa Bilim ve Teknoloji Dergisi. 2020;(19):827-835. doi:https://doi.org/10.31590/ejosat.766623
  • 9. Taylor SJ, Letham B. Forecasting at scale. The American Statistician. 2018;72(1):37-45. doi:https://doi.org/10.7287/peerj.preprints.3190v2
  • 10. Kaya U, Akba F, Medeni İ, Medeni T. Covid-19 öncesi ve sonrasındaki Bitcoin fiyat değişimlerinin Makine Öğrenmesi, zaman serileri analizi ve derin öğrenme yöntemleriyle değerlendirilmesi. Bilişim Teknolojileri Dergisi. 2020;13(3):341-355. doi:https://doi.org/10.17671/gazibtd.648424
  • 11. Napoli AM, Smith-Shain R, Lin T, Baird J. The accuracy of predictive analytics in forecasting emergency department volume before and after onset of COVID-19. Western Journal of Emergency Medicine. 2023;25(1):61. doi:https://doi.org/10.5811/westjem.61059
  • 12. Baccega D, Castagno P, Fernández Anta A, Sereno M. Enhancing COVID-19 forecasting precision through the integration of compartmental models, machine learning and variants. Scientific Reports. 2024;14(1):19220. doi:https://doi.org/10.1038/s41598-024-69660-5
  • 13. Farhat F, Sohail SS, Alam MT, et al. COVID-19 and beyond: leveraging artificial intelligence for enhanced outbreak control. Frontiers in Artificial Intelligence. 2023;6:1266560. doi:https://doi.org/10.3389/frai.2023.1266560
  • 14. ArunKumar K, Kalaga DV, Kumar CMS, Kawaji M, Brenza TM. Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends. Alexandria engineering journal. 2022;61(10):7585-7603. doi:https://doi.org/10.1016/j.aej.2022.01.011
  • 15. Deschepper M, Eeckloo K, Malfait S, Benoit D, Callens S, Vansteelandt S. Prediction of hospital bed capacity during the COVID− 19 pandemic. BMC health services research. 2021;21(1):468. doi:https://doi.org/10.1186/s12913-021-06492-3
  • 16. Ódor G, Karsai M. Epidemic-induced local awareness behavior inferred from surveys and genetic sequence data. Nature Communications. 2025;16(1):4758. doi:https://doi.org/10.48550/arxiv.2406.09983
  • 17. Ali AO, Palanichamy N, Haw SC, Gopal S. Performance Evaluation on COVID-19 Prediction using Machine Learning Models. Journal of Informatics and Web Engineering. 2025;4(2):64-76. doi:https://doi.org/10.33093/jiwe.2025.4.2.5
  • 18. K Abdul Hamid AA, Wan Mohamad Nawi WIA, Lola MS, et al. Improvement of time forecasting models using machine learning for future pandemic applications based on COVID-19 data 2020–2022. Diagnostics. 2023;13(6):1121. doi:https://doi.org/10.3390/diagnostics13061121
  • 19. Zhao D, Zhang R, Zhang H, He S. Prediction of global omicron pandemic using ARIMA, MLR, and Prophet models. Scientific Reports. 2022;12(1):18138. doi: https://doi.org/10.1038/s41598-022-23154-4
  • 20. Nassiri H, Mohammadpour SI, Dahaghin M. How do the smart travel ban policy and intercity travel pattern affect COVID-19 trends? Lessons learned from Iran. Plos one. 2022;17(10):e0276276. doi:https://doi.org/10.1371/journal.pone.0276276
  • 21. Li X, Zhang X. A comparative study of statistical and machine learning models on carbon dioxide emissions prediction of China. Environmental Science and Pollution Research. 2023;30(55):117485-117502. doi:https://doi.org/10.1007/s11356-023-30428-5
  • 22. Wang Y, Xu C, Wu W, et al. Time series analysis of temporal trends in hemorrhagic fever with renal syndrome morbidity rate in China from 2005 to 2019. Scientific Reports. 2020;10(1):9609. doi:https://doi.org/10.1038/s41598-020-66758-4
  • 23. Chen Q, Zheng X, Shi H, et al. Prediction of influenza outbreaks in Fuzhou, China: comparative analysis of forecasting models. BMC Public Health. 2024;24(1):1399. doi:https://doi.org/10.1186/s12889-024-18583-x
  • 24. Zhou L, Zhao P, Wu D, Cheng C, Huang H. Time series model for forecasting the number of new admission inpatients. BMC medical informatics and decision making. 2018;18(1):39. doi:https://doi.org/10.1186/s12911-018-0616-8
  • 25. Jäger WS, Nagler T, Czado C, McCall RT. A statistical simulation method for joint time series of non-stationary hourly wave parameters. Coastal Engineering. 2019;146:14-31. doi:https://doi.org/10.1016/j.coastaleng.2018.11.003
  • 26. Sharma K, Bhalla R, Ganesan G. Time Series Forecasting Using FB-Prophet. 2022:59-65.
  • 27. Zunic E, Korjenic K, Hodzic K, Donko D. Application of facebook's prophet algorithm for successful sales forecasting based on real-world data. arXiv preprint arXiv:200507575. 2020;doi: https://doi.org/10.5121/ijcsit.2020.12203
  • 28. Vo NN, He X, Liu S, Xu G. Deep learning for decision making and the optimization of socially responsible investments and portfolio. Decision Support Systems. 2019;124:113097. doi:https://doi.org/10.1016/j.dss.2019.113097
  • 29. Woltmann L, Deepe J, Hartmann C, Lehner W. evalPM: a framework for evaluating machine learning models for particulate matter prediction. Environmental Monitoring and Assessment. 2023;195(12):1491. doi:https://doi.org/10.1007/s10661-023-11996-y
  • 30. Bandara K, Bergmeir C, Smyl S. Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach. Expert systems with applications. 2020;140:112896. doi:https://doi.org/10.1016/j.eswa.2019.112896
There are 30 citations in total.

Details

Primary Language English
Subjects Health Informatics and Information Systems, Health Counselling, Health Systems, Health and Community Services, Implementation Science and Evaluation, Health Services and Systems (Other)
Journal Section Research Article
Authors

Esra Akaydın Gültürk 0000-0003-0978-3091

Emek Güldoğan 0000-0002-5436-8164

Submission Date October 20, 2025
Acceptance Date December 22, 2025
Publication Date December 31, 2025
Published in Issue Year 2025 Volume: 47 Issue: 4

Cite

AMA Akaydın Gültürk E, Güldoğan E. Investigation of Covid 19 Cases in Sivas Province with Time Series Analysis. CMJ. December 2025;47(4):32-39. doi:10.7197/cmj.1807704