Skip to main content

Table 2 Selection of the univariate ARIMA model

From: Forecasting incidence of hand, foot and mouth disease using BP neural networks in Jiangsu province, China

Model

AIC

BIC

Ljung-Box test

Qa

p

ARIMA(1,0,1)(1,1,0)12

1132.12

1140.43

10.233

0.5272

ARIMA(1,0,0)(1,1,0)12

1135.73

1141.96

17.418

0.1432

ARIMA(2,0,1)(1,1,0)12

1133.77

1144.16

10.187

0.4422

ARIMA(2,0,0)(1,1,0)12

1134.98

1143.29

13.207

0.2944

ARIMA(3,0,1)(1,1,0)12

1134.63

1147.09

9.118

0.4455

ARIMA(3,0,0)(1,1,0)12

1132.76

1143.15

8.6807

0.5811

ARIMA(4,0,1)(1,1,0)12

1136.61

1151.15

8.4172

0.4136

ARIMA(4,0,0)(1,1,0)12

1134.64

1147.11

8.6288

0.4917

  1. aQ denotes the statistics of Ljung-Box test