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

Table 6

MAPE 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	0.077908	0.216183	0.28478	0.031407	4.302802	3.566545	4.780891	3.266033
Test 3	0.005978	0.226541	0.002404	0.002224	0.682114	0.650904	0.325321	0.156613
Train 7	0.107693	0.265915	0.397845	0.070797	8.605604	4.503600	9.561783	3.305931
Test 7	0.007141	0.168689	0.005143	0.00512	0.930147	0.528263	0.446618	0.188684
Train 14	0.135295	0.336248	0.479926	0.108925	8.605604	5.239857	4.780891	3.621599
Test 14	0.008536	0.157663	0.010831	0.0103	1.745331	0.437303	0.482487	0.210358
Train 21	0.143851	0.346251	0.481132	0.142261	7.649426	6.760180	4.780891	3.279957
Test 21	0.00962	0.177791	0.017679	0.016401	1.81934	2.702160	0.487855	0.246182
Train 30	0.153738	0.358279	0.460946	0.176627	1.147414	5.603205	4.780891	2.916246
Test 30	0.010656	0.207111	0.029169	0.026349	1.888606	1.830495	0.456813	0.291251