Skip to main content

Table 2 A series of alternative the SARIMA \((p, 1, q) \times {(P, 1, Q)_{12}}\) models (24 steps ahead prediction)

From: A multivariate multi-step LSTM forecasting model for tuberculosis incidence with model explanation in Liaoning Province, China

(p, q, P, Q)

AIC

BIC

RMSE

MAE

MAPE (%)

sMAPE (%)

(0, 1, 0, 1)

340.64

348.98

0.8471

0.7204

15.9791

14.5258

\(\varvec{(1, 1, 0, 1)}\)

341.49

352.60

0.7634

0.6384

14.1518

13.0495

(0, 2, 0, 1)

341.56

352.68

0.7778

0.6519

14.4502

13.2851

(0, 1, 1, 1)

342.54

353.66

0.8323

0.7083

15.7193

14.3164

(1, 2, 0, 1)

342.83

356.73

0.8040

0.6764

14.9867

13.7548

  1. Bold indicates the best performing model
Back to article page

BMC Infectious Diseases

ISSN: 1471-2334

Contact us