Country Innovation Advantage Ranking: A New Approach and comparison with the Global Innovation Index
收藏资源简介:
We consider a special multi-objective decision-making problem, aiming to identify the ‘best countries’ in terms of their innovative advantage. We propose using rating/ranking-theory methods for the solution to this problem. Moreover, we also show that voting theory methods can be useful to aggregate the results from different ranking methods. For illustrative purposes, our investigation uses a dataset from the Global Innovation Index (GII) 2019. Using different rating/ranking methods, we obtain new alternative ratings/rankings of the innovative advantages of countries. In Particular, the following methods was used as examples: the Buchholz method, Colley method, the Markov-chain method, Perron-Frobenius and geometric mean versions of the analytical hierarchy process (AHP), and the entropy method. The dataset presents the results of our calculations. Namely, the new ratings/rankings of the innovative advantages of countries and their comparison with GII -2019. A detailed description of the calculation methods will be presented in the corresponding publication. The dataset is presented as a sheet of an MS Excel file (.xlsx): which include the following columns: Column #: ID; Column Country: Country name; Column 3 Code: ISO alpha 3 code of country; Column GII-Rating: GII-2019 score Column rE: rating score by the entropy method; Column AHPpf: rating score by the AHP Perron-Frobenius version; Column AHPgm: rating score by the AHP geometric mean version; Column rB: rating score by the Buchholz method; Column rC: rating score by the Colley method; Column Mch: rating score by the Markov-chain method; Column GII-Rank: GII-2019 rank; Column Rank-rE: ranking by the entropy method; Column Rank-AHPpf: ranking by the AHP Perron-Frobenius version; Column Rank-AHPgm: ranking by the AHP geometric mean version; Column Rank-rB: ranking by the Buchholz method; Column Rank-rC: ranking by the Colley method; Column Rank-Mch: ranking by the Markov-chain method; Column AggR: ranking obtained by aggregation previous 6 ranks by the Borda method.
我们针对一类特殊的多目标决策问题展开研究,旨在依据各国的创新优势识别“最佳国家”。我们提出采用评分/排序理论方法求解该问题,同时证明投票理论方法可用于聚合不同排序方法得到的结果。 为便于阐释,本研究采用了2019年全球创新指数(Global Innovation Index, GII)的数据集。通过多种评分/排序方法,我们得到了各国创新优势的全新替代评分与排序结果。本次示例使用的具体方法包括:布赫霍尔茨(Buchholz)方法、科利(Colley)方法、马尔可夫链(Markov-chain)方法、层次分析法(Analytical Hierarchy Process, AHP)的佩龙-弗罗贝尼乌斯(Perron-Frobenius)版本与几何平均版本,以及熵权法。 本数据集呈现了我们的计算成果,即各国创新优势的全新评分与排序结果,并将其与GII-2019的结果进行了对比。关于计算方法的详细阐述将在后续相关出版物中呈现。 本数据集以Microsoft Excel电子表格文件(.xlsx)形式呈现,包含以下列: 1. 第1列:ID; 2. 第2列:Country:国家名称; 3. 第3列:Code:国家的ISO 3字母代码(ISO alpha 3 code); 4. 第4列:GII-Rating:2019年全球创新指数得分; 5. 第5列:rE:熵权法评分; 6. 第6列:AHPpf:层次分析法佩龙-弗罗贝尼乌斯版本评分; 7. 第7列:AHPgm:层次分析法几何平均版本评分; 8. 第8列:rB:布赫霍尔茨方法评分; 9. 第9列:rC:科利方法评分; 10. 第10列:Mch:马尔可夫链方法评分; 11. 第11列:GII-Rank:2019年全球创新指数排名; 12. 第12列:Rank-rE:熵权法排名; 13. 第13列:Rank-AHPpf:层次分析法佩龙-弗罗贝尼乌斯版本排名; 14. 第14列:Rank-AHPgm:层次分析法几何平均版本排名; 15. 第15列:Rank-rB:布赫霍尔茨方法排名; 16. 第16列:Rank-rC:科利方法排名; 17. 第17列:Rank-Mch:马尔可夫链方法排名; 18. 第18列:AggR:通过博达(Borda)法聚合前述6种排名得到的综合排名。




