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

Table 4

RAE values of logistic regression models for fatal cases.

Forecast period (days)	Cumulative cases				Daily cases
	Germany	Japan	South Korea	Ukraine	Germany	Japan	South Korea	Ukraine
Train 3	0.558983	0.748235	0.100892	0.130431	0.982956	0.946141	0.79988	1.233276
Test 3	10.019084	92.770195	3.992647	2.151188	1.450237	3.423077	2.1	0.607143
Train 7	0.761442	0.858548	0.188449	0.264792	1.010998	1.044404	0.76232	1.230431
Test 7	7.11657	19.883806	2.05869	2.056796	1.590134	2.022506	1.510684	0.964646
Train 14	0.939364	1.035772	0.275228	0.410008	1.071022	1.153797	0.760876	1.234803
Test 14	6.073407	7.771993	1.870907	1.948638	1.803453	1.303851	1.431579	0.887464
Train 21	0.98917	1.066849	0.295554	0.548156	1.145123	1.232673	0.802145	1.209536
Test 21	5.33746	5.418942	1.735014	1.822661	1.563513	1.045426	1.450346	0.879166
Train 30	1.062807	1.154789	0.295746	0.677223	1.207509	1.219678	0.84517	1.22699
Test 30	5.473092	4.012622	1.674963	1.726235	1.50333	1.227554	1.405158	0.911704