id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-225347-lnzz2chk	Chakraborty, Tanujit	Nowcasting of COVID-19 confirmed cases: Foundations, trends, and challenges	2020-10-10		.txt	text/plain	10203	585	53	Several statistical and machine learning methods for real-time forecasting of the new and cumulative confirmed cases of COVID-19 are developed to overcome limitations of the epidemiological model approaches and assist public health planning and policy-making [25, 91, 6, 26, 23] . As such, we aim to perform a meaningful data analysis, including the study of time series characteristics, to provide a suitable and comprehensive knowledge foundation for the future step of selecting an apt forecasting method. Five time series COVID-19 datasets for the USA, India, Russia, Brazil, and Peru UK are considered for assessing twenty forecasting models (individual, ensemble, and hybrid). Results for USA COVID-19 data: Among the single models, ARIMA (2, 1, 4) performs best in terms of accuracy metrics for 15-days ahead forecasts. Results for India COVID-19 data: Among the single models, ANN performs best in terms of accuracy metrics for 15-days ahead forecasts.	./cache/cord-225347-lnzz2chk.txt	./txt/cord-225347-lnzz2chk.txt