Deep Ranking Analysis by Power Eigenvectors (DRAPE): A Study. . .
285
of a different initialization of weights on the ranking we considered all the 21
indicators (1) with the equal weights and (2) with weights contribution proportional
to the three wellbeing dimensions. Since the human wellbeing dimension has the
greatest number of indicators, the equal approach is more oriented to humanbased sustainability than the proportional approach which on the other hand
offers a more balanced trade-off between human, environmental and economic
Fig. 6 CPWR ranking and standardized regression coefficients for equal (a and c) and proportional (b and d) weights initializations for Asia-Oceania. CPWR values are represented on the
vertical axis; H stands for the standardized entropy (the coefficient of determination R 2 is equal to
0.95 for the equal weights and 0.94 for the proportional ones)
285
of a different initialization of weights on the ranking we considered all the 21
indicators (1) with the equal weights and (2) with weights contribution proportional
to the three wellbeing dimensions. Since the human wellbeing dimension has the
greatest number of indicators, the equal approach is more oriented to humanbased sustainability than the proportional approach which on the other hand
offers a more balanced trade-off between human, environmental and economic
Fig. 6 CPWR ranking and standardized regression coefficients for equal (a and c) and proportional (b and d) weights initializations for Asia-Oceania. CPWR values are represented on the
vertical axis; H stands for the standardized entropy (the coefficient of determination R 2 is equal to
0.95 for the equal weights and 0.94 for the proportional ones)