Assessing dengue forecasting methods: a comparative study of statistical models and machine learning techniques in Rio de Janeiro, Brazil

Table 2 Accuracy of the statistical models with covariates temperature and humidity at various forecast horizons

Week	SARIMAX			SARIMAX_Lag			VAR			VAR_Lag
Week	MAE	MAPE (%)	RMSE	MAE	MAPE (%)	RMSE	MAE	MAPE (%)	RMSE	MAE	MAPE (%)	RMSE
1	79.24	17.25	128.36	81.80	25.23	125.36	78.10	25.49	123.49	77.74	24.44	116.24
2	114.63	24.27	181.58	114.74	36.96	167.61	118.08	40.70	175.92	108.63	35.11	166.80
3	143.39	30.85	222.82	136.62	39.64	200.87	148.71	48.08	214.39	137.41	43.72	207.91
4	173.52	36.59	278.54	160.28	39.96	246.61	179.45	55.57	267.68	167.53	48.79	259.62
8	316.86	57.84	473.41	279.15	54.86	414.65	313.80	65.08	444.61	295.42	55.03	438.32
12	408.08	67.88	598.16	375.15	73.33	536.71	391.64	64.77	566.66	386.34	62.41	579.80

ISSN: 1349-4147