Forecasting Daily Volume and Acuity of Patients in the Emergency Department

Table 2

MAPE values for each method and 5 different forecasting horizons.

	Forecasting horizon	1	7	14	21	30
Total	MSARIMA	6.23%	12.01%	11.79%	12.29%	11.51%
	SARIMA	6.23%	12.01%	11.79%	12.29%	11.51%
	SS
	SMHW	3.02%	10.80%	10.85%	11.54%	11.11%
VU	MSARIMA	7.19%	17.65%	16.89%	17.23%	17.21%
	SARIMA		17.65%
	SS	11.16%	17.79%	17.36%	18.13%	19.14%
	SMHW	9.40%		16.89%	17.56%	18.18%
U	MSARIMA	3.98%	12.79%	15.71%	14.74%	14.54%
	SARIMA
	SS	7.18%	15.72%	18.04%	16.41%	16.60%
	SMHW	5.57%	14.24%	16.56%	15.50%	15.57%

Models with best fit
MSARIMA: multivariate seasonal autoregressive integrated moving average; SARIMA: seasonal autoregressive integrated moving average; SS: seasonal exponential smoothing; SMHW: seasonal multiplicative Holt-Winters; VU: very urgent; U: urgent.