FORECASTING WITH GROWTH-CURVES - AN EMPIRICAL-COMPARISON

The primary objective of this paper is to compare the forecasting perf ormance of the increasingly wide range of growth curve models. Sevente en models are used to forecast the development of telecommunications m arkets, represented by 25 time series describing telephone penetration in 15 different countries. Forecasting performance is measured by roo t mean square error and mean absolute percentage error over the last 1 0 or 11 years of the series, the model parameters having been fitted o ver the previous 20 years. Note is taken of the convergence of the est imation process, the significance of parameters and the plausibility o f the estimated saturation level. The local logistic, simple logistic and the Gompertz models are shown to significantly outperform more com plex models such as the extended logistic and FLOG models.