In Situ Technological Innovation Diffusion Rate Accuracy Assessment

Abstract

At present, the accuracy of diffusion rate forecasting, at a macro-level, in the research literature, is nonexistent. This research reveals underlying macro-level trends of diffusion rate assessment using historical technological innovation diffusion data to explore the statistical characteristics of diffusion rate percent-error of the Bass and logistic model time stepped through its lifecycle. A quantitative exploratory data analysis (EDA) based approach was employed to uncover underlying macro-perspective patterns and insights on a technological innovation’s forecasted diffusion rate percent-error using the data of 42 matured U.S. consumer technological innovations. An objective of this effort is to determine the statistical characteristics (mean, median, variance, standard deviation, skewness, and kurtosis) of diffusion rate assessment using the Bass and logistic model at various points in a technological innovation’s lifecycle to reveal underlying directional and associative insights. Specifically, this effort explores the development of macro-perspective knowledge on quantifying the forecasting accuracy of a technological innovation’s diffusion rate using partial diffusion data. Developing such insights and a framework for accessing in situ (real-time) a technological innovation’s diffusion rate percent-error would benefit an organization’s decision makers in maximizing gains and minimizing losses. These insights include identifying whether the Bass and logistic models are more likely to overestimate or underestimate a technological innovation’s diffusion rate when assessed at various points in its diffusion lifecycle. Practitioners can use such information to set resource investment strategies and policies based on risk tolerance and the utility of the weighted outcomes via decision theory tools.