Time Series Modeling of Tuberculosis Cases in India from 2017 to 2022 Based on the SARIMA-NNAR Hybrid Model

Table 3

Accuracy parameters of SARIMA, NNAR, and hybrid models of training and testing data set.

Models	Training set			Testing set
	RMSE	MAE	MAPE	RMSE	MAE	MAPE
ARIMA (1, 1, 2)(0, 0, 1) [12]	19104.38	14304.15	09.45	32303.49	22797.31	16.69
NNAR (3, 1, 2) [12]	11566.83	9049.27	05.37	43777.35	36489.51	23.94
Hybrid (SARIMA-NNAR)	13738.97	10369.48	06.68	31591.70	21943.03	15.93