Comparison of Time Series Methods and Machine Learning Algorithms for Forecasting Taiwan Blood Services Foundation’s Blood Supply

Table 2

Error measures obtained under the seven time series models.

Error	Method
	AUTOREG	ARMA	Basic ARIMA	Seasonalized ARIMA	Seasonalized ESM	Multiplicative Holt-Winters	Additive Holt-Winters
MAE	215	449	600	160	158	159	159
MSE	88,031	288,002	577,197	57,235	57,111	57,111	57,189
BIAS	−383	−20,578	754	−5,575	−7,338	−8,507	−15,056
MAPE	94.50	227	224	80	81	81	80