Comparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 Cases

Comparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 Cases | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Comparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 Cases Samyak Prajapati, Aman Swaraj, Ronak Lalwani, Akhil Narwal, Karan Verma, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-493195/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Time series forecasting methods play a critical role in estimating the spread of an epidemic. The coronavirus outbreak of December 2019 has already infected millions all over the world and continues to spread. Just when the curve of the outbreak had started to flatten, many countries have again started to witness a rise in cases which is now being referred to as the 2 nd wave of the pandemic. A thorough analysis of time-series forecasting models is therefore required to equip state authorities and health officials with immediate strategies for future times. Objective: The aims of the study are three-fold: (a) To model the overall trend of the spread; (b) To generate a short-term forecast of 10 days in countries with the highest incidence of confirmed cases (USA, India and Brazil); (c) To quantitatively determine the algorithm that is best suited for precise modelling of the linear and non-linear features of the time series. Comparison: The comparison of forecasting models for the total cumulative cases of each country is carried out by comparing the reported data and the predicted value, and then ranking the algorithms (Prophet, Holt-Winters, LSTM, ARIMA, and ARIMA-NARNN) based on their RMSE, MAE and MAPE values. Result: The hybrid combination of ARIMA and NARNN (Nonlinear Auto-Regression Neural Network) gave the best result among the selected models with a reduced RMSE, which proved to be almost 35.3% better than one of the most prevalent methods of time-series prediction (ARIMA). Conclusion: The r esults demonstrated the efficacy of the hybrid implementation of the ARIMA-NARNN model over other forecasting methods such as Prophet, Holt-Winters, LSTM, and the ARIMA model in encapsulating the linear as well as non-linear patterns of the epidemical datasets. Artificial Intelligence and Machine Learning Hybrid Model Forecasting COVID-19 ARIMA NARNN Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Full Text Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-493195","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":24951866,"identity":"003981cb-5412-4221-9863-aa95a091fd42","order_by":0,"name":"Samyak Prajapati","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIie3RsWrDMBCA4TMCexHNKuOSvsJBhtRg+iwKAXcpWQrBUypjcBY/QCEvkUc4o1Wka7p561Tw6NKhlT3X9tpB/yp96MQBuFz/MD/I80aiYItAAYEPt8iGAzFKbrgusMmSIKwIiHzgs2QpHsuwMekCXyXAQGYH47USm1JHGH7Wut0nfB2A13Zwv5v4i0JLVutoJ4kuKY8LYGEF4nnqFWnJNj49IdWl5qgBIvuXjRoj9ogseTm/m5789IR9z5BcSZMyvPKeUE/86Vd4XYDMEju+HcxctpZ4ZVzhOLk7Hj++umGVZtVm+4clvml97bLDKPkrz16e3Y7L5XK5pvoFakxbq1L6VkAAAAAASUVORK5CYII=","orcid":"","institution":"National Institute of Technology Delhi","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Samyak","middleName":"","lastName":"Prajapati","suffix":""},{"id":24951867,"identity":"746d295d-69b9-4176-aae3-0e95a60ed5a3","order_by":1,"name":"Aman Swaraj","email":"","orcid":"","institution":"National Institute of Technology Delhi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Aman","middleName":"","lastName":"Swaraj","suffix":""},{"id":24951868,"identity":"f6ea624f-c026-4291-b2c0-5241f21a20d1","order_by":2,"name":"Ronak Lalwani","email":"","orcid":"","institution":"National Institute of Technology Delhi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ronak","middleName":"","lastName":"Lalwani","suffix":""},{"id":24951869,"identity":"b1ff72e5-f5ce-44c4-8b45-491b46e36283","order_by":3,"name":"Akhil Narwal","email":"","orcid":"","institution":"National Institute of Technology Delhi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Akhil","middleName":"","lastName":"Narwal","suffix":""},{"id":24951870,"identity":"2992f05c-e626-48d1-a134-eb40599e3296","order_by":4,"name":"Karan Verma","email":"","orcid":"","institution":"National Institute of Technology Delhi","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Karan","middleName":"","lastName":"Verma","suffix":""},{"id":24951871,"identity":"d2c0cac8-6c48-4735-9911-7ed147bda35b","order_by":5,"name":"Ghanshyam Singh","email":"","orcid":"","institution":"Malaviya National Institute of Technology Jaipur","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ghanshyam","middleName":"","lastName":"Singh","suffix":""},{"id":24951872,"identity":"9cb782e8-24cf-457c-afa3-a0e9191d8b68","order_by":6,"name":"Ashok Kumar","email":"","orcid":"","institution":"Government Mahila Engineering College, Ajmer","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Ashok","middleName":"","lastName":"Kumar","suffix":""}],"badges":[],"createdAt":"2021-05-04 18:00:09","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":true,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false,"coiExplicitlySet":false},"doi":"10.21203/rs.3.rs-493195/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-493195/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":8800011,"identity":"ab6cee4f-0809-4c7f-b910-eef13ad7aaa3","added_by":"auto","created_at":"2021-05-05 13:20:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":155395,"visible":true,"origin":"","legend":"Second wave of COVID-19 rising in European\ncountries respectively from top: Germany, Italy and UK\n(February 1 to November 14, 2020).","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/bc1c8177a87cdc8eb7d72bba.png"},{"id":8800055,"identity":"a025b9db-0692-4743-ae56-974398ad49a8","added_by":"auto","created_at":"2021-05-05 13:20:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":28682,"visible":true,"origin":"","legend":"Flowchart indicating the various steps involved for\nforecasting","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/512c5ca96e21c5c4421b1f8d.png"},{"id":8800060,"identity":"0eddd6ef-c72b-4b21-bc89-36ddd97a239e","added_by":"auto","created_at":"2021-05-05 13:20:43","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":40001,"visible":true,"origin":"","legend":"Pictorial Representation of an LSTM Node [40]","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/2b400a95159080ce4849c66f.png"},{"id":8800013,"identity":"63c6d9bf-d13d-4aeb-bb9a-6f01698f753d","added_by":"auto","created_at":"2021-05-05 13:20:33","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":358532,"visible":true,"origin":"","legend":"Pictorial representation of the process for the ARIMANARNN Hybrid model [41]","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/df809b854c8564228bb7e9ce.png"},{"id":8800252,"identity":"a5c8b872-c3c2-44ce-973f-aa900e80c3d6","added_by":"auto","created_at":"2021-05-05 13:23:32","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":146860,"visible":true,"origin":"","legend":"Graphical Illustration of the forecasts for\nIndia respectively on 6th May - 15th May, 21st July- 30th July\nand 1st August- 10th August from top\n","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/c47d84eade7769c0c2815241.png"},{"id":8800015,"identity":"441d96e9-a5bb-41b6-bee4-473f39bbf61a","added_by":"auto","created_at":"2021-05-05 13:20:33","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":77596,"visible":true,"origin":"","legend":"Graphical Illustration of the forecasts for USA\nand Brazil respectively from top (between 6 - 15th May, 2020)","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1/1cea9bc3a656f76d26fabf2f.png"},{"id":13628367,"identity":"876478ef-a7b3-46ed-a697-0227f8a4e2f0","added_by":"auto","created_at":"2021-09-17 07:59:15","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":950974,"visible":true,"origin":"","legend":"","description":"","filename":"FINALPAPERresearchsquare.pdf","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1_covered.pdf"},{"id":8800545,"identity":"5c579af9-aab6-406c-8431-33747e07ad7d","added_by":"auto","created_at":"2021-05-05 13:26:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":730181,"visible":true,"origin":"","legend":"","description":"","filename":"FINALPAPERresearchsquare.pdf","url":"https://assets-eu.researchsquare.com/files/rs-493195/v1_stamped.pdf"}],"financialInterests":"","formattedTitle":"\u003cp\u003eComparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 Cases\u003c/p\u003e","fulltext":[{"header":"Full Text","content":"This preprint is available for \u003ca href='/article/rs-493195/latest.pdf' target='_blank'\u003edownload as a PDF\u003c/a\u003e."}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"National Institute of Technology Delhi","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Hybrid Model, Forecasting, COVID-19, ARIMA, NARNN","lastPublishedDoi":"10.21203/rs.3.rs-493195/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-493195/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Time series forecasting methods play a critical role in estimating the spread of an epidemic. The coronavirus outbreak of December 2019 has already infected millions all over the world and continues to spread. Just when the curve of the outbreak had started to flatten, many countries have again started to witness a rise in cases which is now being referred to as the 2\u003csup\u003end\u003c/sup\u003e wave of the pandemic. A thorough analysis of time-series forecasting models is therefore required to equip state authorities and health officials with immediate strategies for future times.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eObjective: \u003c/strong\u003eThe aims of the study are three-fold: (a) To model the overall trend of the spread; (b) To generate a short-term forecast of 10 days in countries with the highest incidence of confirmed cases (USA, India and Brazil); (c) To quantitatively determine the algorithm that is best suited for precise modelling of the linear and non-linear features of the time series.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eComparison: \u003c/strong\u003eThe comparison of forecasting models for the total cumulative cases of each country is carried out by comparing the reported data and the predicted value, and then ranking the algorithms (Prophet, Holt-Winters, LSTM, ARIMA, and ARIMA-NARNN) based on their RMSE, MAE and MAPE values.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eResult:\u003c/strong\u003e The hybrid combination of ARIMA and NARNN (Nonlinear Auto-Regression Neural Network) gave the best result among the selected models with a reduced RMSE, which proved to be almost 35.3% better than one of the most prevalent methods of time-series prediction (ARIMA).\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eThe\u003cstrong\u003e r\u003c/strong\u003eesults demonstrated the efficacy of the hybrid implementation of the ARIMA-NARNN model over other forecasting methods such as Prophet, Holt-Winters, LSTM, and the ARIMA model in encapsulating the linear as well as non-linear patterns of the epidemical datasets.\u003c/p\u003e","manuscriptTitle":"Comparison of Traditional and Hybrid Time Series Models for Forecasting COVID-19 Cases","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-05-05 13:19:04","doi":"10.21203/rs.3.rs-493195/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"145965fc-6641-4faa-9043-a6f9e76e79ad","owner":[],"postedDate":"May 5th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":4083455,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2021-05-05T13:19:04+00:00","versionOfRecord":[],"versionCreatedAt":"2021-05-05 13:19:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-493195","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-493195","identity":"rs-493195","version":["v1"]},"buildId":"ApUGefWb6u5IBVtyqm6d5","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}