Skip to main content

Table 3 Cross-Validation of Seasonal Naive Point Forecasts

From: Forecasting the daily demand for emergency medical ambulances in England and Wales: a benchmark model and external validation

Horizon (days)

MASE

sMAPE

RMSE

7

0.94 (0.35)

3.48 (1.19)

96.91 (37.11)

14

1.06 (0.39)

3.94 (1.32)

109.71 (41.88)

21

1.12 (0.41)

4.12 (1.38)

115.38 (42.94)

28

1.14 (0.40)

4.19 (1.33)

118.34 (42.92)

35

1.18 (0.40)

4.35 (1.34)

123.26 (42.33)

42

1.23 (0.40)

4.51 (1.33)

128.11 (40.34)

49

1.25 (0.39)

4.61 (1.32)

131.40 (39.79)

56

1.27 (0.39)

4.67 (1.33)

133.95 (40.20)

63

1.30 (0.40)

4.77 (1.36)

136.59 (40.30)

70

1.31 (0.37)

4.82 (1.26)

138.70 (37.25)

77

1.33 (0.34)

4.88 (1.18)

140.73 (34.49)

84

1.34 (0.35)

4.94 (1.21)

143.04 (34.07)

365

1.49 (0.46)

5.51 (1.64)

155.29 (38.64)

  1. Figures are forecast horizon (days), Mean Absolute Scaled Error (standard deviation), symmetric Mean Absolute Percentage Error (standard deviation) and Root Mean Squared Error (standard deviation) by forecast horizon (\(n = 27\) folds)