A data envelopment analysis and local partial least squares approach for identifying the optimal innovation policy direction
| Autor: | Dionisis Philippas, Alexandros Leontitsis, Robin C. Sickles, Panagiotis Tziogkidis |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
050210 logistics & transportation
021103 operations research Information Systems and Management General Computer Science Computer science 05 social sciences 0211 other engineering and technologies 02 engineering and technology Management Science and Operations Research Industrial and Manufacturing Engineering National innovation system Modeling and Simulation 0502 economics and business Partial least squares regression Data envelopment analysis Econometrics Global Innovation Index Diversity (business) |
| Zdroj: | European Journal of Operational Research. 285:1011-1024 |
| ISSN: | 0377-2217 |
| Popis: | The paper proposes a novel two-step approach that evaluates countries’ innovation efficiency and their responsiveness to expansions in their innovation inputs, while addressing shortcomings associated with composite indicators. Based on our evaluations, we propose innovation policies tailored to take into account the diverse economic environments of the many countries in our study. Applying multidirectional efficiency analysis on data from the Global Innovation Index, we obtain separate efficiency scores for each innovation input and output. We then estimate different sensitivities for each country, by applying partial least squares on explanatory and response matrices which are determined by the nearest neighbors of the country under consideration. The findings reveal substantial asymmetries with respect to innovation efficiencies and sensitivities, which is indicative of the diversity of national innovation systems. Considering these two dimensions in combination, we outline three policy directions that can be followed, offering a platform for better-informed decision-making. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect