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

Table 4 Benchmark results by forecast horizon

Horizon (days)	MASE	Coverage 80%	Coverage 95%
7	0.66 (0.64 - 0.69)	0.846 (0.829 - 0.864)	0.961 (0.952 - 0.970)
14	0.67 (0.65 - 0.69)	0.847 (0.835 - 0.860)	0.964 (0.958 - 0.970)
21	0.67 (0.65 - 0.68)	0.850 (0.839 - 0.860)	0.964 (0.959 - 0.970)
28	0.67 (0.66 - 0.69)	0.849 (0.840 - 0.858)	0.965 (0.961 - 0.969)
35	0.67 (0.66 - 0.69)	0.849 (0.841 - 0.858)	0.965 (0.961 - 0.969)
42	0.68 (0.66 - 0.69)	0.847 (0.839 - 0.855)	0.965 (0.962 - 0.969)
49	0.68 (0.67 - 0.69)	0.847 (0.839 - 0.855)	0.965 (0.962 - 0.969)
56	0.68 (0.67 - 0.70)	0.847 (0.839 - 0.854)	0.966 (0.963 - 0.969)
63	0.69 (0.68 - 0.70)	0.846 (0.839 - 0.854)	0.966 (0.964 - 0.969)
70	0.69 (0.68 - 0.70)	0.845 (0.838 - 0.853)	0.967 (0.964 - 0.969)
77	0.70 (0.68 - 0.71)	0.844 (0.837 - 0.852)	0.967 (0.964 - 0.969)
84	0.70 (0.69 - 0.71)	0.843 (0.835 - 0.851)	0.966 (0.963 - 0.969)

Mean (95% CI) MASE and 80, 95% prediction interval coverage. Figures pool all simulated forecasts from all seven sub-regions (\(n = 287\); 41 folds per sub-region)

ISSN: 1472-6947