id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
aijbe-50	Yeliz , Zoğal	PROPHET MODEL'S EFFICIENCY IN SHORT-TERM COVID-19 CUMULATIVE CASE PROJECTIONS: G7 COUNTRIES	2022	10	.pdf	application/pdf	5743	278	57	The model has three different components 𝑦�(𝑡�) = 𝑔�(𝑡�) + 𝑠�(𝑡�) + ℎ(𝑡�) + 𝜀�𝑡� (1) Here, 𝑦�(𝑡�) is the number of cases, 𝑔�(𝑡�) is the trend, that is, the trend function that captures non-periodic changes in the time series, 𝑠�(𝑡�) is the seasonality, that is, the part of the time series that captures periodic (weekly or yearly) changes, ℎ(𝑡�) is the part that represents holidays that occur at irregular intervals, and 𝜀�𝑡� is the error term that includes specific changes that cannot be covered or handled by the model [17]. The study used ARIMA model to forecast the trend of the COVID-19 outbreak in Italy, Spain, and France, which are the countries that were most affected by the pandemic in Europe, using data from the period of February 21 to April 15, 2020.	cache/aijbe-50.pdf	txt/aijbe-50.txt