Technological Forecasting—Model Selection, Model Stability, and Combining Models

Abstract

The paper identifies 29 models that the literature suggests are appropriate for technological forecasting. These models are divided into three classes according to the timing of the point of inflexion in the innovation or substitution process. Faced with a given data set and such a choice, the issue of model selection needs to be addressed. Evidence used to aid model selection is drawn from measures of model fit and model stability. An analysis of the forecasting performance of these models using simulated data sets shows that it is easier to identify a class of possible models rather than the ‘best’ model. This leads to the combining of model forecasts. The performance of the combined forecasts appears promising with a tendency to outperform the individual component models.

Exploring the dynamics of journal citations: Modelling with s-curves

“…The difficulty of predicting future citations demonstrated by this example reflects very well Meade andIslam's (1998, p. 1116) comments about technological forecasting using s-curves:…”

“…(Martino, 1983). However, there are many curves that could be used 2 and in fact Meade and Islam (1998) identify 29 different ones. Meade and Islam classify their models into three classes -symmetric, asymmetric and flexible depending on behaviour around the point of inflection.…”

“…It is a very flexible distribution whose probability function can take a variety of shapes from right skew through normality to left skew. It is non-symmetric and flexible in Meade and Islam's (1998) terms.…”

“…It is similar to the Weibull in taking a variety of shapes from the exponential to the normal. It is nonsymmetric and flexible in Meade and Islam's (1998) terms.…”

Forecasting and operational research: a review

“…Many univariate alternatives (summarized in Meade and Islam, 2006) have been proposed, including the Gompertz curve (which has the same S-shaped form as the logistic), as well as Harvey's (1984) and Meade's (1985) contributions in JORS, all of which have shown comparative empirical success. But there is apparently no best function and again, combining may be the best approach (Meade and Islam, 1998). Shore and Benson-Karhi (2007) are more optimistic that selection can be productive.…”