Modeling Weekly Covid Data in Europe and Sri Lanka: Time Series Approach

PROCEEDINGS OF THE SLIIT INTERNATIONAL CONFERENCE ON ADVANCEMENTS IN SCIENCES AND HUMANITIES [SICASH] Pub Date : 2022-10-11 DOI:10.54389/fxwm6962

J. Jayakody

{"title":"Modeling Weekly Covid Data in Europe and Sri Lanka: Time Series Approach","authors":"J. Jayakody","doi":"10.54389/fxwm6962","DOIUrl":null,"url":null,"abstract":"Novel Corona Virus, commonly known as COVID-19 has become a global threat affecting more than 200 countries up to date. Still a vaccine that can assure of hundred percent prevention has not been discovered. All the countries are currently following WHO guidelines such as lockdowns and social distancing. This study was conducted to develop ARIMA models for COVID-19 data in Europe and Sri Lanka and validate the models. For both these regions, number of COVID-19 cases were collected considering for a period of one year in which the first real wave happened. ACF and PACF plots were used to identify the stationarity, and out of the results possible ARIMA models were developed for the two regions separately. For Europe, the best fitted model was ARIMA (0, 2, 1) and for Sri Lanka, the best fitted model was ARIMA (1,1,0). The models were evaluated using AIC criteria. The errors of the models were found to be white noise. The forecasted values that were obtained from the model showed an increase of cases in Europe and a constant flow in Sri Lanka. Keywords: ARIMA Models, Covid-19, Forecasting","PeriodicalId":112882,"journal":{"name":"PROCEEDINGS OF THE SLIIT INTERNATIONAL CONFERENCE ON ADVANCEMENTS IN SCIENCES AND HUMANITIES [SICASH]","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"PROCEEDINGS OF THE SLIIT INTERNATIONAL CONFERENCE ON ADVANCEMENTS IN SCIENCES AND HUMANITIES [SICASH]","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.54389/fxwm6962","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Novel Corona Virus, commonly known as COVID-19 has become a global threat affecting more than 200 countries up to date. Still a vaccine that can assure of hundred percent prevention has not been discovered. All the countries are currently following WHO guidelines such as lockdowns and social distancing. This study was conducted to develop ARIMA models for COVID-19 data in Europe and Sri Lanka and validate the models. For both these regions, number of COVID-19 cases were collected considering for a period of one year in which the first real wave happened. ACF and PACF plots were used to identify the stationarity, and out of the results possible ARIMA models were developed for the two regions separately. For Europe, the best fitted model was ARIMA (0, 2, 1) and for Sri Lanka, the best fitted model was ARIMA (1,1,0). The models were evaluated using AIC criteria. The errors of the models were found to be white noise. The forecasted values that were obtained from the model showed an increase of cases in Europe and a constant flow in Sri Lanka. Keywords: ARIMA Models, Covid-19, Forecasting

查看原文本刊更多论文

欧洲和斯里兰卡每周新冠肺炎数据建模:时间序列方法

新型冠状病毒，俗称COVID-19，迄今已成为影响200多个国家的全球性威胁。然而，一种能够保证百分之百预防的疫苗还没有被发现。所有国家目前都在遵循世卫组织的指导方针，如封锁和保持社交距离。本研究旨在为欧洲和斯里兰卡的COVID-19数据开发ARIMA模型并对模型进行验证。对于这两个地区，收集的COVID-19病例数考虑到第一次真正浪潮发生的一年时间。利用ACF和PACF图确定平稳性，并根据结果分别建立了两个地区的ARIMA模型。对于欧洲，最佳拟合模型为ARIMA(0,2,1)，对于斯里兰卡，最佳拟合模型为ARIMA(1,1,0)。采用AIC标准对模型进行评价。发现模型误差为白噪声。从该模型获得的预测值显示，欧洲病例增加，斯里兰卡病例持续增加。关键词:ARIMA模型，Covid-19，预测

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

PROCEEDINGS OF THE SLIIT INTERNATIONAL CONFERENCE ON ADVANCEMENTS IN SCIENCES AND HUMANITIES [SICASH]

自引率

0.00%

发文量