PUB - Publikationen an der Universität Bielefeld

Based on data for 14 consumer electronic products and using the Gompertz curve as a benchmark, NN (2011) carried out a “more extensive testing” of the utility-based sales forecasting approach proposed by Decker and Gnibba-Yukawa (2010). However, at least from a practical point of view, the plausibility of their testing framework regarding the market potential m is not unquestionable. The results presented in this article suggest that the utility-based approach could achieve more accurate forecasts when applying a more reasonable range for m, rather than varying it between 2 and 15 times the cumulative sales by the end of year 7 as proposed by NN (2011). Furthermore, the predictive accuracy of the Gompertz curve may significantly improve for short time series if m, as opposed to their suggestion, is estimated exogenously. With an appropriate m, the utility-based and the Gompertz curve approach tend to perform on a par. The overall forecast accuracy can be further improved if the individual forecasts of both approaches are combined to one prediction. Combining the outcomes of different forecasting approaches promises to be an effective strategy for early sales forecasting in high-technology markets, rather than relying on one presumably “best” model.