مقایسه مدل‌ها در پیش‌بینی موارد تجمعی بستری و فوت کووید-19 (مطالعه موردی: شهرستان بهاباد)
محورهای موضوعی : -اقتصاد بهداشت و درمانمحمد حسین کریمی زارچی 1 , داود شیشه بری 2
1 - کارشناسی‌ارشد مهندسی صنایع، دانشکده فنی و مهندسی، دانشگاه یزد، یزد، ایران
2 - دانشیار، گروه مهندسی صنایع، دانشکده فنی و مهندسی، دانشگاه یزد، یزد، ایران
کلید واژه: سری زمانی, مدلسازی آماری, پاندمیک, کووید-19, پیش‌بینی,
چکیده مقاله :
مقدمه: بیماری کووید-19، یک بیماری تنفسی است که در اثر سندرم تنفسی حاد کرونا ویروس-2 ایجاد شده است. پیش‌بینی تعداد موارد جدید و مرگومیر می‌تواند گام مفیدی در پیش‌بینی هزینه‌ها و امکانات مورد نیاز در آینده باشد. هدف از این مطالعه مدلسازی، مقایسه عملکرد مدل‌ها و پیش‌بینی موارد جدید بستری و مرگومیر در آینده نزدیک است. روش پژوهش: در این مقاله 9 تکنیک پیش‌بینی بر روی داده‌های کووید-19 شهرستان بهاباد استان یزد تحت آزمایش قرار گرفت و با استفاده از معیارهای ارزیابی میانگین مربعات خطا (MSE)، جذر میانگین مربعات خطا (RMSE)، میانگین قدر مطلق خطا (MAE) و میانگین درصد قدرمطلق خطا (MAPE) مدل‌ها باهم مقایسه شدند. یافته‌ها: نتایج تحلیل نشان داد، بهترین مدل با توجه به معیارهای ارزیابی مذکور برای پیش‌بینی موارد تجمعی بستری کووید-19 مدل هموارسازی اسپلاین مکعبی و برای موارد تجمعی فوت مدل رگرسیون KNN می‌باشد. هم‌چنین مدل شبکه‌های عصبی اتورگرسیو و مدل تتا برای موارد بستری و برای موارد فوت مدل شبکه‌های عصبی اتورگرسیو دارای بدترین عملکرد را در میان دیگر مدل‌ها دارا می‌باشد. نتیجه گیری: این مطالعه می‌تواند درک مناسبی از روند شیوع بیماری کووید-19 در این منطقه ارائه کند تا با اتخاذ اقدامات احتیاطی و تدوین سیاست های مناسب بتوان به نحو احسن از این بیماری عبور کرد. هم‌چنین برخلاف مطالعات دیگر این مطالعه، از 9 تکنیک متفاوت و مقایسه آن‌ها، استفاده کرده است که به نوبه خود ضریب اطمینان را در تصمیم‌گیری بالا برده است. هم‌چنین نکته‌ای که حائز اهمیت می‌باشد این است که باید داده‌ها در زمان واقعی بروز شوند.
Introduction: Coronavirus disease 2019 is a respiratory disease caused by acute respiratory syndrome coronavirus-2. Forecasting the number of new cases and deaths during todays can be a useful step in predicting the costs and facilities needed in the future. This study aims to modeling, comparing the performance of models, and predict new cases and deaths efficiently in the future. Methods: In this article nine popular forecasting techniques are tested on the data of Covid-19 in Bahabad city as a case study. Using the evaluation criteria of mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), and the mean absolute percentage of error (MAPE) of the models are compared. Results: The results of the analysis showed that the best model according to the evaluation criteria for forecasting cumulative cases of hospitalization of Covid-19 is the cubic spline smoothing model, and cumulative cases of death, KNN regression model. Also, autoregressive neural network and theta models for hospitalization cases, and for death cases, autoregressive neural network model has the worst performance among other models. Conclusion: This study can provide a proper understanding of the spread of covid-19 disease in this region so that by taking precautionary measures and formulating appropriate policies, this epidemic can be effectively overcome. Also, unlike other studies, this study uses 9 different techniques and their comparison, which in turn increases the confidence factor in decision making. Also, an important point is that the data should be updated in real time.
1- Cucinotta D, Vanelli M. WHO declares COVID-19 a pandemic. Acta bio medica: Ateneiparmensis, 2020; 91(1): 157.
2- Cheng ZJ, Shan J. 2019 Novel coronavirus: where we are and what we know. Infection, 2020 Apr; 48: 155-63.
3- Zhang X, Liu Y, Yang M, Zhang T, Young AA, Li X. Comparative study of four time series methods in forecasting typhoid fever incidence in China. PloS one, 2013 May 1; 8(5): e63116.
4- Olanrewaju SO, Ojo EO, Oguntade ES. Time Series Analysis on Reported Cases of Tuberculosis in Minna Niger State Nigeria. Open Journal of Statistics, 2020 May 8; 10(3): 412-30.
5- Nayak MS, Narayan KA. Forecasting dengue fever incidence using ARIMA analysis. International Journal of Collaborative Research on Internal Medicine & Public Health, 2019; 11(6): 924-32.
6- Wu W, An SY, Guan P, Huang DS, Zhou BS. Time series analysis of human brucellosis in mainland China by using Elman and Jordan recurrent neural networks. BMC infectious diseases, 2019 Dec; 19(1): 1-1.
7- Golkhandan A, Sahraei S. The Prediction of Iran's Per Capita Health Expenditures up to 2041 HorizonUsing the Genetic and Particle Swarm Optimization Algorithms. Journal of Healthcare Management, 2019; 9(4): 66-53. [In Persian]
8- Ceylan Z. Estimation of COVID-19 prevalence in Italy, Spain, and France. Science of The Total Environment, 2020 Aug 10; 729: 138817.
9- Bayyurt L, Bayyurt B. Forecasting of COVID-19 cases and deaths using ARIMA models. Medrxiv, 2020 Apr 22: 2020-04.
10- Khan FM, Gupta R. ARIMA and NAR based prediction model for time series analysis of COVID-19 cases in India. Journal of Safety Science and Resilience, 2020 Sep 1; 1(1): 12-8.
11- Alassafi MO, Jarrah M, Alotaibi R. Time series predicting of COVID-19 based on deep learning. Neurocomputing, 2022 Jan 11; 468: 335-44.
12- Mukhairez HH, Alaff AJ. Short-term Forecasting of COVID-19. Computational Intelligence for COVID-19 and Future Pandemics: Emerging Applications and Strategies, 2022: 257-66.22.
13- Li C, Sampene AK, Agyeman FO, Robert B, Ayisi AL. Forecasting the severity of COVID-19 pandemic amidst the emerging SARS-CoV-2 variants: adoption of ARIMA model. Computational and Mathematical Methods in Medicine, 2022 Jan 13; 2022.
14- Kufel T. ARIMA-based forecasting of the dynamics of confirmed Covid-19 cases for selected European countries. Equilibrium. Quarterly Journal of Economics and Economic Policy. 2020; 15(2): 181-204.
15- Martínez F, Frías MP, Charte F, Rivera AJ. Time Series Forecasting with KNN in R: the tsfknn Package. R J. 2019 Dec 1; 11(2): 229.
16- Nikolopoulos K, Assimakopoulos V, Bougioukos N, Litsa A, Petropoulos F. The theta model: An essential forecasting tool for supply chain planning. InAdvances in Automation and Robotics, Vol. 2: Selected Papers from the 2011 International Conference on Automation and Robotics (ICAR 2011), Dubai, December 1–2; 2011: 431-437. Springer Berlin Heidelberg.
17- Wen Q, Gao J, Song X, Sun L, Xu H, Zhu S. RobustSTL: A robust seasonal-trend decomposition algorithm for long time series. InProceedings of the AAAI Conference on Artificial Intelligence, 2019 Jul 17; 33(01): 5409-5416.
18- Wang Y. Smoothing splines: methods and applications. CRC press; 2011 Jun 22.
19- De Livera AM, Hyndman RJ, Snyder RD. Forecasting time series with complex seasonal patterns using exponential smoothing. Journal of the American statistical association, 2011 Dec 1; 106(496): 1513-27.
_||_1- Cucinotta D, Vanelli M. WHO declares COVID-19 a pandemic. Acta bio medica: Ateneiparmensis, 2020; 91(1): 157.
2- Cheng ZJ, Shan J. 2019 Novel coronavirus: where we are and what we know. Infection, 2020 Apr; 48: 155-63.
3- Zhang X, Liu Y, Yang M, Zhang T, Young AA, Li X. Comparative study of four time series methods in forecasting typhoid fever incidence in China. PloS one, 2013 May 1; 8(5): e63116.
4- Olanrewaju SO, Ojo EO, Oguntade ES. Time Series Analysis on Reported Cases of Tuberculosis in Minna Niger State Nigeria. Open Journal of Statistics, 2020 May 8; 10(3): 412-30.
5- Nayak MS, Narayan KA. Forecasting dengue fever incidence using ARIMA analysis. International Journal of Collaborative Research on Internal Medicine & Public Health, 2019; 11(6): 924-32.
6- Wu W, An SY, Guan P, Huang DS, Zhou BS. Time series analysis of human brucellosis in mainland China by using Elman and Jordan recurrent neural networks. BMC infectious diseases, 2019 Dec; 19(1): 1-1.
7- Golkhandan A, Sahraei S. The Prediction of Iran's Per Capita Health Expenditures up to 2041 HorizonUsing the Genetic and Particle Swarm Optimization Algorithms. Journal of Healthcare Management, 2019; 9(4): 66-53. [In Persian]
8- Ceylan Z. Estimation of COVID-19 prevalence in Italy, Spain, and France. Science of The Total Environment, 2020 Aug 10; 729: 138817.
9- Bayyurt L, Bayyurt B. Forecasting of COVID-19 cases and deaths using ARIMA models. Medrxiv, 2020 Apr 22: 2020-04.
10- Khan FM, Gupta R. ARIMA and NAR based prediction model for time series analysis of COVID-19 cases in India. Journal of Safety Science and Resilience, 2020 Sep 1; 1(1): 12-8.
11- Alassafi MO, Jarrah M, Alotaibi R. Time series predicting of COVID-19 based on deep learning. Neurocomputing, 2022 Jan 11; 468: 335-44.
12- Mukhairez HH, Alaff AJ. Short-term Forecasting of COVID-19. Computational Intelligence for COVID-19 and Future Pandemics: Emerging Applications and Strategies, 2022: 257-66.22.
13- Li C, Sampene AK, Agyeman FO, Robert B, Ayisi AL. Forecasting the severity of COVID-19 pandemic amidst the emerging SARS-CoV-2 variants: adoption of ARIMA model. Computational and Mathematical Methods in Medicine, 2022 Jan 13; 2022.
14- Kufel T. ARIMA-based forecasting of the dynamics of confirmed Covid-19 cases for selected European countries. Equilibrium. Quarterly Journal of Economics and Economic Policy. 2020; 15(2): 181-204.
15- Martínez F, Frías MP, Charte F, Rivera AJ. Time Series Forecasting with KNN in R: the tsfknn Package. R J. 2019 Dec 1; 11(2): 229.
16- Nikolopoulos K, Assimakopoulos V, Bougioukos N, Litsa A, Petropoulos F. The theta model: An essential forecasting tool for supply chain planning. InAdvances in Automation and Robotics, Vol. 2: Selected Papers from the 2011 International Conference on Automation and Robotics (ICAR 2011), Dubai, December 1–2; 2011: 431-437. Springer Berlin Heidelberg.
17- Wen Q, Gao J, Song X, Sun L, Xu H, Zhu S. RobustSTL: A robust seasonal-trend decomposition algorithm for long time series. InProceedings of the AAAI Conference on Artificial Intelligence, 2019 Jul 17; 33(01): 5409-5416.
18- Wang Y. Smoothing splines: methods and applications. CRC press; 2011 Jun 22.
19- De Livera AM, Hyndman RJ, Snyder RD. Forecasting time series with complex seasonal patterns using exponential smoothing. Journal of the American statistical association, 2011 Dec 1; 106(496): 1513-27.