provides a possible means for classification of technological substitutions
.
Specifically, the functional linear regression model developed in this study has been able to correctly identify the expected mode of substitution for 19 out of 20 technologies with data available during the emergence phase, and has done so using patent indicator counts of the number of non-corporates and the number of cited references by priority year. These patent data dimensions can in turn be associated with technological and scientific production respectively, in line with the predictions made about requirements for demonstrating the conditions for presumptive technological substitutions based on the work of Constant in section \ref{585124}. Whilst this pairing of patent indicators demonstrated the most robust statistical and out-of-sample performance of any of the patent indicator subsets considered as mode predictors from the emergence stage, this does not prove that these are the only indicators capable of doing so. As discussed in section \ref{311620}, the possibility of orthogonality has not been ruled out with regards to the other patent indicators shown in Table \ref{table:bibliometric_indicators}. However, these two dimensions are in good agreement with the technological anomaly arguments put forward by Constant, and so were felt to be reasonable for forming the basis of the technology classification model that has been developed.
based on the two principal classes considered from literature evidence
Conclusions from statistical ranking and functional data analysis
Drawing from the ideas of Constant and preceding technological substitution studies this paper has attempted to explore the possibility of classification of two principle modes of substitution relationships between
Whilst it has not been possible to detail all elements of this analysis
It was postulated in section
Technology classification model is built on the assumptions given in section \ref{585124}.
In this regard the technology classification system is based on technology profile dimensions relating to both scientific and technological progress.
If both of these components are missing, the functional linear regression model defaults to a prediction of technological substitution by functional failure (i.e. there does not appear to be any opportunity for adoption).
Limited number of technologies considered (cross-validation would benefit from a more diverse group of technologies - see section \ref{258858})
Two principal modes of technology substitution examined from literature and technology adoption case studies: reactive and presumptive
Statistical analysis of patent indicator time series provides a possible means for classification of substitutions: for the datasets considered measures of the number of cited references and the involvement of non-corporate entities by year during the emergence phase were found to provide a good indication of the mode of substitution and performed consistently well in statistical ranking of predictive capability
Sensitivity of technology adoption to chosen modelling parameters
Whilst statistical approaches are well-suited to detecting underlying correlations in historical and experimental datasets, this on it's own does not provide a detailed understanding of the causation behind associated events. Equally, statistical methods are not generally well suited to predicting disruptive events and complex interactions, with other simulation techniques such as System Dynamics and Agent Based Modelling performing better in these areas. Accordingly, in order to identify causation effects and test the sensitivity of technological substitution patterns to variability arising from real-world socio-technical features not captured in simple bibliometric indicators (such as the influence of competition and economic effects), the fitted regression model is evaluated in a real-time system dynamics environment.
- Preliminary adoption data appears to show a distinction between those technologies arising as a result of technological failure, and those arising based on a presumptive technological leap (to be confirmed)
- From available patent data indicators ‘cited patents’ and ‘cited references’ do seem to be able to provide a means of determining the mode of adoption during the emergence phase of the Technology Life Cycle – these two indicators are normally taken to correspond to the rates of technological and scientific progress respectively in a given field
- Patent indicator subset selected for use in model building based on ranking exercise does appear to provide the basis for a statistically significant technology classification model
- Functional data analysis appears to provide valid method to build a technology classification model based on specific Technology Life Cycle stages
- High-dimensional functional model found to have the highest significance based on F-ratio statistics comparison
- Permutation testing of the functional linear regression analysis also suggests that the model built is sensitive to the order of the technology time series being considered (particularly in the high-dimensional case), so this relationship would appear to be based on the specifics of the individual technology curves considered, and does not appear to be occurring by chance
- Comparison of functional linear regression vs. functional principal components analysis: conclusions?