Multivariate Covid-19 Forecasting with Vaccinations as a factor: the case of India and USA
2022 IEEE Region 10 Symposium, TENSYMP 2022
; 2022.
Article
in English
| Scopus | ID: covidwho-2052090
ABSTRACT
Accurate forecasting of Covid-19 case load is essential to ensure healthcare system preparedness in all countries due to highly infectious strains like Omicron. Although many countries have started vaccination drives, forecasting of case numbers predominantly hasn't accounted for vaccinations. This paper investigates whether multivariate models that include vaccinations as a factor such as VAR, VARIMA and Multivariate LSTM, perform better than their univariate counterparts AR, ARIMA and Univariate LSTM, at forecasting daily case numbers. Both long-term and short-term forecast accuracies of the models have been compared using the RMSE, MAE and MAPE metrics. This study is conducted in the context of cases and vaccinations in India and USA up to January 2022 to find out the relative effect of the rate of vaccination on case load and contrast the situations in the two countries. © 2022 IEEE.
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Topics:
Vaccines
Language:
English
Journal:
2022 IEEE Region 10 Symposium, TENSYMP 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS