Skip to main content

Table 2 Forecast errors for the various SARIMA models

From: Time series modelling to forecast prehospital EMS demand for diabetic emergencies

Model MAE MSE MAPE
Overall
(0,1,0,12) 22.96 758.39 4.23%
 (0,1,1,12) 24.88 870.29 4.63%
 (1,1,0,12) 25.94 942.60 4.79%
 (1,1,1,12) 24.54 862.56 4.56%
 (0,1,2,12) 24.33 849.05 4.52%
 (2,1,0,12) 23.79 794.36 4.41%
 (2,1,1,12) 26.31 866.99 4.81%
 (1,1,2,12) 26.09 909.21 4.81%
Hypoglycaemia
 (0,1,0,12) 38.12 2026.34 11.58%
(0,1,1,12) 25.50 941.22 7.57%
 (1,1,0,12) 26.85 1055.83 8.04%
 (1,1,1,12) 26.63 951.10 7.75%
 (0,1,2,12) 27.34 994.96 7.90%
 (2,1,0,12) 25.17 906.32 7.35%
 (2,1,1,12) 27.84 1037.25 7.90%
 (1,1,2,12) 27.73 1063.02 8.06%
Hyperglycaemia
 (0,1,0,12) 31.26 1412.35 14.74%
(0,1,1,12) 12.94 254.31 6.26%
 (1,1,0,12) 15.49 321.36 7.43%
 (1,1,1,12) 13.22 269.20 6.40%
 (0,1,2,12) 13.38 281.07 6.49%
 (2,1,0,12) 15.35 337.68 7.41%
 (2,1,1,12) 15.24 348.73 7.41%
 (1,1,2,12) 13.78 300.21 6.67%
Female
 (0,1,0,12) 26.21 918.92 10.81%
 (0,1,1,12) 20.68 536.97 8.64%
 (1,1,0,12) 22.40 632.35 9.31%
 (1,1,1,12) 21.45 582.90 9.04%
 (0,1,2,12) 20.76 547.93 8.75%
(2,1,0,12) 14.78 360.32 6.12%
 (2,1,1,12) 17.93 414.51 7.47%
 (1,1,2,12) 19.72 482.18 8.25%
Male
 (0,1,0,12) 24.97 872.20 7.97%
 (0,1,1,12) 20.12 590.74 6.38%
 (1,1,0,12) 22.34 746.52 7.03%
(1,1,1,12) 18.85 618.37 5.99%
 (0,1,2,12) 18.97 634.50 6.03%
 (2,1,0,12) 18.78 437.23 6.11%
 (2,1,1,12) 20.70 605.84 6.56%
 (1,1,2,12) 20.54 706.99 6.51%
  1. MAE (Mean Absolute Error), MSE (Mean Square Error), MAPE (Mean Absolute Percentage Error)
  2. Bold text indicates chosen model