Effective Utilization of Data for Predicting COVID-19 Dynamics: An Exploration through Machine Learning Models

Table 1

MAE values of logistic regression models for confirmed cases.

Forecast period (days)	Cumulative cases				Daily cases
	Germany	Japan	South Korea	Ukraine	Germany	Japan	South Korea	Ukraine
Train 3	857718.6	286153.96	108432	55505.78	48276.21	18819.15	77411.48	5451.58
Test 3	772078.67	1587192	159047.67	33767.33	39960.33	31894.67	13655	14126
Train 7	1162616.09	445712.26	213582.58	136885.78	48782.19	20648.93	65870.27	5232.94
Test 7	988841.57	2865681.57	382687.57	94103.43	28426.29	23223.57	21234.43	15376.14
Train 14	1444834.85	551208.48	546825.8	241053.19	50805	22077.31	66760.47	6004.26
Test 14	1183490.57	3308088.5	1601407.64	279010.07	24988.43	27008.71	32216.71	15638.57
Train 21	1971303.35	607461.92	798955.03	291836.94	53295.33	23092.32	74409.74	6541.27
Test 21	1341054.1	3797773.29	2219878.71	421748.81	23132.67	30210.1	38164.95	14651.62
Train 30	1952622.63	688047.48	1134894.41	354217.3	53247.71	18295.2	74542.03	6966.61
Test 30	1534048	4897131.16	2920608.16	565430.74	24795.03	39843.06	32511.26	11527.39