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

Table 9 Average width of the uncertainty intervals of models using cases alone

Week	Statistical models				Machine learning models					Ensemble
Week	AR(1)	MA(1)	ARIMA	ETS	SVM	RF	XGBoost	LSTM	Prophet	LSTM_ARIMA
1	362.85	356.35	395.18	335.26	412.05	411.82	478.14	343.56	858.24	333.64
2	575.24	563.12	575.84	480.79	568.55	615.88	696.32	514.02	965.86	468.90
3	716.57	757.53	741.74	603.25	712.56	796.76	903.85	683.67	1038.23	661.30
4	897.53	935.67	970.22	803.09	856.58	977.07	1045.36	852.74	1094.66	841.25
8	1440.92	1486.77	1505.23	1220.31	1353.92	1511.77	1548.26	1465.60	1346.44	1483.38
12	1881.42	1877.10	1973.47	1880.60	1678.33	1856.88	1901.31	1837.23	1579.96	1881.55

ISSN: 1349-4147