Table 3: Accuracy of the models of patient arrivals with a 1-hour forecasting interval.

Department Model MAE MAPE MASE
In Out In Out In Out
ED1
Regression 1.61 1.62 51 58 0.76 0.76
ARIMA (2,0,2)×(2,0,1)168 1.60 1.59 50 58 0.76 0.75
Naïve model 2.12 2.19 69 76 1.00 1.03
ED2
Regression 1.81 1.75 48 47 0.75 0.73
ARIMA (2,0,4)×(1,0,1)168 1.77 1.72 47 49 0.74 0.72
Naïve model 2.40 2.36 65 66 1.00 0.99
ED3
Regression 1.67 1.70 50 54 0.76 0.77
ARIMA (2,0,2)×(1,0,1)168 1.64 1.66 48 56 0.74 0.75
Naïve model 2.21 2.22 68 75 1.00 1.00
ED4
Regression 1.83 1.82 48 48 0.75 0.75
ARIMA (3,0,3)×(1,0,1)168 1.79 1.82 47 51 0.73 0.75
Naïve model 2.43 2.31 66 60 1.00 0.95

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).