{"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