A Novel Method for Objective Criterion Weighting: The Neyman Chi-Square Distance Approach (NCDA)

Authors : Furkan Fahri Altıntaş

Pages : 444-469

Doi:10.7240/jeps.1782364

View : 81 | Download : 220

Publication Date : 2025-12-23

Article Type : Research Paper

Abstract :In the context of Multi-Criteria Decision-Making (MCDM), most objective weighting techniques tend to focus either solely on internal variability (e.g., ENTROPY, SD, SVP, LOPCOW) or exclusively on external structural effects (e.g., MEREC). Although the CRITIC method considers both dimensions, it remains constrained by its dependence on linear associations and parametric assumptions—particularly the assumption of normality. Such methodological limitations can undermine the reliability of decision outcomes and constitute the central motivation of this study. To address this limitation, the study introduces an innovative framework termed the Neyman Chi-Square Distance Approach (NCDA). NCDA is a non-parametric method that simultaneously accounts for both internal variation and external structural divergence. Standard deviation captures internal variability, while Neyman Chi-Square Distance (NCD) quantifies external differences without relying on correlation assumptions. Empirical analyses and case studies reveal that NCDA demonstrates strong robustness against perturbations in the weighting process, as validated through sensitivity analysis. Comparative evaluations confirm the method’s high consistency with well-established approaches, while simulation experiments highlight its capacity to produce balanced and stable weight distributions. Overall, NCDA provides an original and methodologically rigorous contribution to the field by directly integrating distributional distance analysis into the weighting process of decision-making models. This dual-perspective framework not only strengthens the theoretical foundations but also enhances the adaptability and reliability of MCDM applications across complex and heterogeneous data environments.
Keywords : NCD, Kriter ağırlığı, ÇKKV, NCDM

ORIGINAL ARTICLE URL