Table 6.

Performance of GBM in the prediction of recovered and active cases in Maharashtra, India

Performance parameter	Distribution
	Gamma	Quintile	Poisson	Laplace	Huber	Tweedie	Gaussian
Recovered cases
MSE	21,005.42	24,869.43	691.80	25,379.27	21,081.05	7094.25	3889.94
RMSE	144.93	157.70	26.30	159.30	145.19	84.22	62.36
MAE	102.83	89.67	18.27	96.70	68.90	44.64	39.55
MRD	12.17	44.83	−1975.29	96.70	9993.07	2.58	3889.94
R2	0.52	0.43	0.98	0.42	0.52	0.84	0.91
Active cases
MSE	816,125.8	1,572,996	643,718.7	613,581.1	271,677.3	201,587.4	109,033.4
RMSE	903.39	1254.19	802.32	783.31	521.22	448.98	330.2021
MAE	599.62	891.48	344.46	545.42	355.50	280.64	252.3535
MRD	16.74	445.74	−21,429.08	545.42	154,812.5	9.82	109,033.4
R2	0.56	0.16	0.65	0.67	0.85	0.89	0.94

Note: R2 (coefficient of determination), MSE (mean square error), RMSE (root mean square error), MAE (mean average error), and MRD (mean residual deviance)