On the Efficiency of Grey Modeling in Early-Stage Technological Diffusion Forecasting

On the Efficiency of Grey Modeling in Early-Stage Technological Diffusion Forecasting

Charisios Christodoulos (National and Kapodistrian University of Athens, Athens, Greece), Christos Michalakelis (Harokopio University of Athens, Athens, Greece) and Thomas Sphicopoulos (National and Kapodistrian University of Athens, Athens, Greece)
Copyright: © 2015 |Pages: 11
DOI: 10.4018/IJTD.2015040101
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The issue of how to obtain an accurate short-term forecast in the beginning stage of the technological diffusion is of great importance for policy makers, researchers and managers. Time-series forecasting has been noticeably neglected in the specific research area due to the prerequisite of having enough data in order to create a time-series. In this paper, Grey modeling is examined as an alternative tool for technology diffusion forecasting in the early diffusion process, where the commonly used aggregate diffusion models usually fail to deliver accurate forecasts. Grey modeling is a unique time-series methodology that requires only a few data points in order to make a forecast. The GM(1,1) model is tested against a classic aggregate diffusion model, the Gompertz model, using only the first four data of mobile broadband diffusion to make an one-step-ahead prediction. The results in the EU15 countries reveal that the Grey model outperforms the Gompertz model in every case, thus stimulating new research guidelines in terms of combinations of the two approaches and further investigation of the value of Grey modeling in the specific area.
Article Preview

Introduction

One of the key objectives of technology management is accurate diffusion forecasting, even at the beginning of the diffusion process, where the available data are not sufficient for application of well-known models. As the rapid pace of technological innovation allows heterogeneous technologies to coexist and converge, emerging innovations have been introduced to overcome the limitations of existing ones and to meet consumers' requirements. Accurate early forecasting allows policy makers, researchers and managers to control the changeable market and enhance competition. This research topic is always up-to-date, and recent research papers in the area stimulate the scientific interest (Goodwin et al, 2014, Nguimkeu, 2014, Shi et al, 2014, Lin 2013). It is obvious that an accurate and easily applicable method for forecasting the diffusion of innovations would be an extremely beneficial tool for companies, especially when they need to estimate the diffusion of new-to-the-market products. It is crucial for a company seeking sustainable competitive advantage to anticipate future developments on its markets. The usual techniques used for this purpose are divided into two categories: Qualitative and Quantitative techniques (Fildes & Kumar, 2002, Gruber and Verboven, 2001). In the world of research, there are two general approaches to gathering and reporting information: qualitative and quantitative approaches. The qualitative approach to research is focused on understanding a phenomenon from a closer perspective. The quantitative approach tends to approximate phenomena from a larger number of individuals using survey methods.

Qualitative techniques in the specific research area include:

  • Delphi method converges answers from a panel of experts

  • Scenario planning envisions multiple possible futures and their implications

  • Qualitative diffusion models describe a bell curve of innovators, early adopters, early majority, late majority, and laggards and the process of how innovations diffuse from one group to the next.

Quantitative techniques in the specific research area include:

  • “S-curves,” such as the Bass model which provides a mathematical model based on a population of innovators and imitators, the Logistic model and the Gompertz model, based on biological population dynamics.

  • Causal techniques use regression testing to identify key variables that determine a specific technology’s penetration.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 9: 4 Issues (2018): 1 Released, 3 Forthcoming
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing