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

Table 14

MAE values of support vector regression models for fatal cases.

Forecast period (days)	Cumulative cases				Daily cases
	Germany	Japan	South Korea	Ukraine	Germany	Japan	South Korea	Ukraine
Train 3	399.16	90.81	72.14	617.96	50.06	14.44	28.87	40.12
Test 3	102.69	180.45	20.51	64.21	41.55	21.76	5.51	21.52
Train 7	681.17	171.72	214.52	1067.67	50.84	15.75	30.78	44.9
Test 7	225.59	461.32	21.5	38.32	27.67	14.07	3.79	15.48
Train 14	1202.87	321.98	641.75	1778.12	60.34	19.37	33.74	58.43
Test 14	352.16	1085.13	22.18	136.92	24.66	36.1	9.06	24.67
Train 21	1723.38	476.29	1037.56	2673.69	68.13	24.05	39.41	67.3
Test 21	439.12	1787.47	20.5	394.72	22.38	39.28	11.32	42.76
Train 30	2471.49	743.6	1612.14	3814.64	77.89	27.76	41.99	89.43
Test 30	496.3	2963.69	19.63	985.68	22.33	37.67	13.84	57.79