Table 5: Accuracy of the models of ED occupancy with a 1-hour forecasting interval.
| Department |
Model |
MAE |
MAPE |
MASE |
| In |
Out |
In |
Out |
In |
Out |
| ED1 |
|
|
|
|
|
|
|
|
Regression |
5.46 |
5.46 |
116 |
117 |
0.73 |
0.73 |
|
ARIMA (2,0,1)×(1,0,1)168
|
2.45 |
2.57 |
72 |
73 |
0.33 |
0.34 |
|
Naïve model |
7.52 |
7.50 |
192 |
191 |
1.00 |
1.00 |
| ED2 |
|
|
|
|
|
|
|
|
Regression |
5.47 |
5.46 |
115 |
115 |
0.75 |
0.75 |
|
ARIMA (2,0,1)×(1,0,1)168
|
2.60 |
2.67 |
68 |
69 |
0.36 |
0.37 |
|
Naïve model |
7.29 |
7.29 |
170 |
170 |
1.00 |
1.00 |
| ED3 |
|
|
|
|
|
|
|
|
Regression |
5.41 |
5.40 |
115 |
115 |
0.73 |
0.72 |
|
ARIMA (2,0,1)×(1,0,1)168
|
2.61 |
2.68 |
68 |
70 |
0.35 |
0.36 |
|
Naïve model |
7.46 |
7.42 |
177 |
177 |
1.00 |
1.00 |
| ED4 |
|
|
|
|
|
|
|
|
Regression |
17.93 |
17.71 |
137 |
136 |
0.75 |
0.74 |
|
ARIMA (2,1,4)×(2,0,1)168
|
5.28 |
5.43 |
100 |
101 |
0.22 |
0.23 |
|
Naïve model |
24.00 |
23.63 |
266 |
264 |
1.00 |
0.99 |
Note: MAE = mean absolute error, MAPE = mean absolute percentage error, MASE = mean absolute scaled error, In = in sample (i.e., calculated over the data used in building the model: January 2012 – December 2014), Out = out of sample (i.e., calculated over the forecasted data: January 2015).