Forecast Evaluation in Call Centers: Combined Forecasts, Flexible Loss Functions and Economic Criteria
33 Pages Posted: 4 May 2011
Date Written: March 22, 2011
Accurate forecasts of incoming calls are crucial to optimal staffing decisions in call centers. This paper evaluates a wide range of models and forecast combination techniques by means of statistical and economic criteria. Relative to the previous literature, this paper is novel in several respects. In particular, the statistical evaluation of competing models is carried out by using a flexible loss function as input to pairwise and joint forecast diagnostic checks. Informative rankings across alternative single models and different groups of models are obtained. Moreover, models are evaluated from the perspective of a manager, who needs reliable forecasts to dimension the call center. Money metrics of forecasting performance are computed, which are based on the economic value of information and the certainty equivalent.
Keywords: Combining forecasts, Decision making, Loss function, Seasonality, Telecommunications forecasting
JEL Classification: C22, C44, C53, D81, L96
Suggested Citation: Suggested Citation