Comparing the performance of time series models with or without meteorological factors in predicting incident pulmonary tuberculosis in eastern China

Table 1 The monthly number of pulmonary tuberculosis cases in the three cities in 2018 predicted by the ARIMA, ARIMAX, and RNN models

Month	Xuzhou city				Nantong city				Wuxi city
Month	Observation	ARIMA	ARIMAX	RNN	Observation	ARIMA	ARIMAX	RNN	Observation	ARIMA	ARIMAX	RNN
January	237	239	227	258	193	145	177	194	158	137	135	141
February	188	259	249	249	177	149	163	202	112	143	141	170
March	300	323	304	227	261	216	258	209	194	174	175	177
April	273	291	277	285	221	195	202	219	188	177	184	139
May	271	288	283	282	251	196	212	206	206	178	181	194
June	273	293	286	285	230	196	204	224	183	202	204	188
July	229	242	237	250	204	167	186	200	218	183	187	193
August	248	257	237	242	166	195	205	218	206	198	200	214
September	169	255	247	215	170	189	170	207	203	195	199	192
October	208	206	238	192	151	134	125	178	150	160	170	177
November	193	209	243	180	149	146	150	183	191	183	181	199
December	241	206	240	201	162	195	212	190	177	178	181	179

ARIMA autoregressive integrated moving average, ARIMAX autoregressive integrated moving average with exogenous variables, RNN recurrent neural network

ISSN: 2049-9957