A New Index to Measure the Consistency of Reciprocal Matrix

MING ZHENG,YI SHEN

DOI: https://doi.org/10.3969/j.issn.1007-6093.2000.01.009

2000-01-01

Abstract:In this paper, we provide a new index that can be used to measure the consistency of reciprocal matrix when geometric mean is derived as the weight vector solution. And this index is compared with the one constructed by the largest eigenvalue of the reciprocal matrix.

What problem does this paper attempt to address?

Geometric consistency index for interval pairwise comparison matrices

Jiancheng Tu,Zhibin Wu,Jiuping Xu

DOI: https://doi.org/10.1080/01605682.2022.2075803

IF: 3.6

2022-05-26

Journal of the Operational Research Society

Abstract:Interval pairwise comparison matrices are widely accepted for practical decision making problems when the decision maker is unable to provide an exact judgment on the alternatives. However, as measuring the preference consistencies in pairwise comparison decision making problems is important, this paper proposes a new interval pairwise comparison matrix consistency measure, the average geometric consistency index, that assumes that the preference in a given interval follows the lognormal distribution. This geometric consistency measure accounts for the interval boundaries and uncertainties. As it is often difficult to rationally rank alternatives when interval pairwise comparison matrices are highly uncertain and/or inconsistent, we propose an optimization model to reduce the inconsistencies of these matrices while minimizing information loss and controlling uncertainties. An interval priority vector is derived to rank the alternatives. The feasibility and efficiency of the models are demonstrated using numerical examples.

management,operations research & management science
CONVERGENCE AND STABILITY OF ITERATIVELY REWEIGHTED LEAST SQUARES FOR LOW-RANK MATRIXRECOVERY

Yun Cai,Song Li

DOI: https://doi.org/10.3934/ipi.2017030

2017-01-01

Inverse Problems and Imaging

Abstract:In this paper, we study the theoretical properties of iteratively reweighted least squares algorithm for recovering a matrix (IRLS-M for short) from noisy linear measurements. The IRLS-M was proposed by Fornasier et al. (2011) [ 17 ] for solving nuclear norm minimization and by Mohan et al. (2012) [ 31 ] for solving Schatten- \begin{document}$p$\end{document} (quasi) norm minimization ( \begin{document}$0 ) in noiseless case, based on the iteratively reweighted least squares algorithm for sparse signal recovery (IRLS for short) (Daubechies et al., 2010) [ 15 ], and numerical experiments have been given to show its efficiency (Fornasier et al. and Mohan et al.) [ 17 ], [ 31 ]. In this paper, we focus on providing convergence and stability analysis of iteratively reweighted least squares algorithm for low-rank matrix recovery in the presence of noise. The convergence of IRLS-M is proved strictly for all \begin{document}$0 . Furthermore, when the measurement map \begin{document}$\mathcal{A}$\end{document} satisfies the matrix restricted isometry property (M-RIP for short), we show that the IRLS-M is stable for \begin{document}$0 . Specially, when \begin{document}$p=1$\end{document} , we prove that the M-RIP constant \begin{document}$δ_{2r} is sufficient for IRLS-M to recover an unknown (approximately) low rank matrix with an error that is proportional to the noise level. The simplicity of IRLS-M, along with the theoretical guarantees provided in this paper, make a compelling case for its adoption as a standard tool for low rank matrix recovery.
Further Discussion on C.R.

ZHENG Ming,SHEN Yi,HUANG Zhi-bin

DOI: https://doi.org/10.3969/j.issn.1005-2542.2001.01.017

2001-01-01

Abstract:When applying AHP to make decision, decision makers always use C.R. to measure the consistency of a reciprocal matrix. If C.R. ≤10%, then the reciprocal matrix is thought to be consistent, otherwise it is refused. But does it make sense to use C.R., the ratio of the C.I. to an average C.I. (known as R.I.) computed from 500 n×n random reciprocal matrices, to examine the consistency?And why should the upper bound of C.R. be 10%?The purpose of this paper is to explain the statistical meaning of using the consistency index C.R. fr om the point of estimate error control. We also discuss the rationality to use the 10% cut-off rule.
Geometric mean method for judgement matrices: Formulas for errors

I.L. Tomashevskii

DOI: https://doi.org/10.48550/arXiv.1410.0823

2014-10-03

Abstract:The geometric mean method (GMM) and the eigenvector method (EM) are well-known approaches to deriving information from pairwise comparison matrices in decision making processes. However, the original algorithms of these methods are logically incomplete and have significant drawbacks: their actual numerical errors are unknown and their reliability is doubted by different rank reversal phenomena including the GMM-EM ones. Recently (Tomashevskii 2015: Eur J Oper Res 240:774-780), the actual numerical errors were found for EM, and it was shown that all EM rank reversal phenomena have the same cause and are eliminated when the errors are taken into account. In this paper the similar ap-proach is used for GMM: we associate GMM with some standard measuring procedure, ana-lyze this procedure from the viewpoint of measurement theory, and find the actual GMM errors. We show that the GMM and the GMM-EM rank reversal phenomena are eliminated when the GMM and the EM errors are taken into account. The GMM decision support tool, which has all components of a standard measuring tool, is composed of pairwise comparisons as an initial measuring procedure, GMM as a data processor, and the obtained formulas for GMM errors as an error indicator. This tool is analogous to the EM decision support tool received earlier in the above paper. It is shown that the EM and the GMM versions of the tool are equally suitable to measure and rank any comparable elements with positive numerical values. We also analyze the Geometric Consistency Index usually used in the original GMM to measure of the inconsistency and to accept or reject an inconsistent pairwise comparison matrix, and show that this index is not an acceptable GMM error indicator.

Optimization and Control
Consistency and Consensus Improving Methods for Pairwise Comparison Matrices Based on Abelian Linearly Ordered Group

Meimei Xia,Jian Chen

DOI: https://doi.org/10.1016/j.fss.2014.07.019

IF: 4.462

2015-01-01

Fuzzy Sets and Systems

Abstract:The aim of a valued pairwise comparison matrix is to derive the priority structure over a set of criteria (or alternatives) in decision making. The consistency and consensus of a pairwise comparison matrix should be measured and improved to avoid a misleading priority structure. The basic entries of a pairwise comparison matrix can be described in different forms; hence, different consistency and consensus methods should be developed for different types of matrices. To provide a general framework, the pairwise comparison matrix based on Abelian linearly ordered group is first introduced. A consistency index is defined by constructing the nearest consistent pairwise comparison matrix from an inconsistent one, and two consistency improving methods are introduced. A group pairwise comparison matrix is derived, a consensus index of individual pairwise comparison matrices is defined and two consensus improving methods are developed by introducing a general aggregation operator based on Abelian linearly ordered group. The proposed consistency and consensus methods are convergent and can provide a general framework for existing methods.
Efficiency analysis for the Perron vector of a reciprocal matrix

Susana Furtado,Charles Johnson

2024-04-22

Abstract:In prioritization schemes, based on pairwise comparisons, such as the Analytical Hierarchy Process, it is necessary to extract a cardinal ranking vector from a reciprocal matrix that is unlikely to be consistent. It is natural to choose such a vector only from efficient ones. One of the most used ranking methods employs the (right) Perron eigenvector of the reciprocal matrix as the vector of weights. It is known that the Perron vector may not be efficient. Here, we focus on extending arbitrary reciprocal matrices and show, constructively, that two different extensions of any fixed size always exist for which the Perron vector is inefficient and for which it is efficient, with the following exception. If B is consistent, any reciprocal matrix obtained from B by adding one row and one column has efficient Perron vector. As a consequence of our results, we obtain families of reciprocal matrices for which the Perron vector is inefficient. These include known classes of such matrices and many more. We also characterize the 4-by-4 reciprocal matrices with inefficient Perron vector. Some prior results are generalized or completed.

Combinatorics
New inconsistency indicators for incomplete pairwise comparisons matrices

Jacek Szybowski,Konrad Kułakowski,Anna Prusak

DOI: https://doi.org/10.48550/arXiv.1912.00505

2019-12-07

Abstract:We introduce two new inconsistency measures for the incomplete pairwise comparisons matrices and show several examples of their calculation. We also carry out a comparative analysis of the new inconsistency indices with the existing ones based on the Monte Carlo simulation.

Information Theory
Efficiency of the convex hull of the columns of certain triple perturbed consistent matrices

Susana Furtado,Charles Johnson

2024-07-09

Abstract:In decision making a weight vector is often obtained from a reciprocal matrix A that gives pairwise comparisons among n alternatives. The weight vector should be chosen from among efficient vectors for A. Since the reciprocal matrix is usually not consistent, there is no unique way of obtaining such a vector. It is known that all weighted geometric means of the columns of A are efficient for A. In particular, any column and the standard geometric mean of the columns are efficient, the latter being an often used weight vector. Here we focus on the study of the efficiency of the vectors in the (algebraic) convex hull of the columns of A. This set contains the (right) Perron eigenvector of A, a classical proposal for the weight vector, and the Perron eigenvector of AA^{T} (the right singular vector of A), recently proposed as an alternative. We consider reciprocal matrices A obtained from a consistent matrix C by modifying at most three pairs of reciprocal entries contained in a 4-by-4 principal submatrix of C. For such matrices, we give necessary and sufficient conditions for all vectors in the convex hull of the columns to be efficient. In particular, this generalizes the known sufficient conditions for the efficiency of the Perron vector. Numerical examples comparing the performance of efficient convex combinations of the columns and weighted geometric means of the columns are provided.

Combinatorics
Efficient Vectors for Block Perturbed Consistent Matrices

Susana Furtado,Charles Johnson

DOI: https://doi.org/10.1137/23m1580310

IF: 1.908

2024-02-10

SIAM Journal on Matrix Analysis and Applications

Abstract:SIAM Journal on Matrix Analysis and Applications, Volume 45, Issue 1, Page 601-618, March 2024. In prioritization schemes, based on pairwise comparisons, such as the analytical hierarchy process, it is important to extract a cardinal ranking vector from a reciprocal matrix that is unlikely to be consistent. It is natural to choose such a vector only from efficient ones. Recently, a method to generate inductively all efficient vectors for any reciprocal matrix has been discovered. Here we focus on the study of efficient vectors for a reciprocal matrix that is a block perturbation of a consistent matrix in the sense that it is obtained from a consistent matrix by modifying entries only in a proper principal submatrix. We determine an explicit class of efficient vectors for such matrices. Based on this, we give a description of all the efficient vectors in the 3-by-3 block perturbed case. In addition, we give sufficient conditions for the right Perron eigenvector of such matrices to be efficient and provide examples in which efficiency does not occur. Also, we consider a certain type of constant block perturbed consistent matrices, for which we may construct a class of efficient vectors, and demonstrate the efficiency of the Perron eigenvector. Appropriate examples are provided throughout.

mathematics, applied
Optimizing consistency and consensus in group decision making based on relative projection between multiplicative reciprocal matrices

Zu-Lin Liu,Fang Liu,Jia-Wei Zhang,Witold Pedrycz

DOI: https://doi.org/10.1016/j.eswa.2023.119948

IF: 8.5

2023-03-29

Expert Systems with Applications

Abstract:Individual consistency and group consensus are the two protagonists in group decision-making (GDM). The former is considered as the logical basis of decision information. The latter is regarded as the guarantee of obtaining a widely accepted solution. In this study, two novel indexes for measuring consistency and consensus levels are defined using relative projection between multiplicative reciprocal matrices (MRMs) originating in Analytic Hierarchy Process (AHP). They are called the relative projection consistency index (RPCI) and the geometric relative projection consensus index (GRPCI), respectively. In order to test acceptable consistency of MRMs, the thresholds of RPCI are determined using the simulation method. Then, an optimization method for improving consistency of MRMs is proposed. By considering acceptable consistency of individual MRMs, a model of optimizing group consensus is further established by minimizing the GRPCI. Finally, a novel GDM algorithm is elaborated on, where the degrees of individual consistency and group consensus are optimized. A case study is carried out to illustrate the proposed models, where the Gaussian quantum behavior particle swarm optimization (GQPSO) algorithm is used to solve the constructed optimization models. Comparison with these previous studies reveals that a novel perspective is offered to simultaneously measure consistency degree of MRMs and group consensus level in GDM. The constructed optimization models are effective for improving consistency and reaching consensus in GDM with MRMs.

computer science, artificial intelligence,engineering, electrical & electronic,operations research & management science
Analysis of natural cardinal ranking vectors for pairwise comparisons and the universal efficiency of the Perron geometric mean

S. Furtado,C. R. Johnson

2024-09-02

Abstract:In models using pair-wise (ratio) comparisons among alternatives, a cardinal ranking vector should be deduced from a reciprocal matrix. The right Perron eigenvector (RP) was traditionally used, though several other options have emerged. We consider some alternatives, mostly new, namely the entry-wise reciprocal of the left Perron vector (LP), the left singular vector (LS), the entry-wise reciprocal of the right singular vector (RS), the arithmetic and geometric means of RP and LP (AP and GP), and of LS and RS (AS and GS). (The ranking vector AS was proposed by Gass and Rapcsák (2004)). All 8 of these vectors produce the natural vector in the consistent case. We compare them empirically, in terms of efficiency, for random matrices, as a function of the number of alternatives. It turns out that the vector GP is universily efficient, and this fact is proven. The vector GS performs better that the remaining 6 vectors. We show that, for reciprocal matrices obtained from consistent matrices by modifying one column and the corresponding row, all 8 vectors are efficient. Moreover, the cone generated by the columns is efficient.

Combinatorics
A gradient method for inconsistency reduction of pairwise comparisons matrices

Jean-Pierre Magnot,Jiří Mazurek,Viera Čerňanová

DOI: https://doi.org/10.48550/arXiv.2108.04034

2022-07-16

Abstract:We investigate an application of a mathematically robust minimization method -- the gradient method -- to the consistencization problem of a pairwise comparisons (PC) matrix. Our approach sheds new light on the notion of a priority vector and leads naturally to the definition of instant priority vectors. We describe a sample family of inconsistency indicators based on various ways of taking an average value, which extends the inconsistency indicator based on the "$\sup$"- norm. We apply this family of inconsistency indicators both for additive and multiplicative PC matrices to show that the choice of various inconsistency indicators lead to non-equivalent consistencization procedures.

Rings and Algebras,Numerical Analysis,Optimization and Control,Statistics Theory
A New Method of Improving the Consistency of Fuzzy Judgement Matrices Based on Compatibility

Hua-you CHEN,Chun-lin LIU

DOI: https://doi.org/10.3969/j.issn.1001-7402.2005.02.024

2005-01-01

Abstract:Abstrcat Fuzzy judgement matrix is a kind of important preference information form provided by (decision) maker. Based on the complementary characteristics of fuzzy judgement matrix, a (compatibility) index is proposed in this paper. The relations between compatibility and consistency of (fuzzy) judgement matrices are also studied. Then a deviation matrix is defined which can reflect the (difference) of fuzzy judgement matrix and its characteristic matrix. A new method of improving the (consistency) of fuzzy judgement matrices is given. Finally, an example shows the feasibility and (effectiveness) of this method.
A VARIANT OF THE REFINED ARNOLDI METHOD FOR LARGE SCALE MATRIX EIGENPROBLEMS

贾仲孝,陈桂芝

DOI: https://doi.org/10.3969/j.issn.1000-3266.2003.02.003

2003-01-01

Abstract:The refined Arnoldi method proposed by Jia is used for computing some eigen-pairs of large matrices. In contrast to the Arnoldi method, the fundamental dif-ference is that the refined method seeks certain refined Ritz vectors, which aredifferent from the Ritz vectors obtained by the Arnoldi method, from a projection space with minimal residuals to approximate the desired eigenvectors. In com-parison with the Ritz vectors, the refined Ritz vectors are guaranteed to converge theoretically and can converge much faster numerically. In this paper we propose to replace the Ritz values, obtained by the Arnoldi method with respect to a Krylovsubspace, by the ones obtained with respect to the subspace spanned by the refined Ritz vectors. We discuss how to compute these new approximations cheaply and reliably. Theoretical error bounds between the original Ritz values and the new Ritz values are established. Finally, we present a variant of the refined Arnoldi al-gorithm for an augmented Krylov subspace and discuss restarting issue. Numerical results confirm efficiency of the new algorithm.
Algorithms for improving consistency or consensus of reciprocal [0,1]-valued preference relations

Meimei Xia,Zeshui Xu,Jian Chen

DOI: https://doi.org/10.1016/j.fss.2012.09.016

IF: 4.462

2013-01-01

Fuzzy Sets and Systems

Abstract:We investigate the consistency and consensus of reciprocal [0,1]-valued preference relations (also called fuzzy preference relations by many authors) based on the multiplicative consistency property, which is an important issue in fuzzy set theory. An algorithm is first developed to improve the consistency level of a reciprocal [0,1]-valued preference relation, and the corresponding algorithm for the incomplete reciprocal [0,1]-valued preference relation is also developed. We further propose the consensus improving algorithms for individual reciprocal [0,1]-valued preference relations or incomplete ones. The convergence and robustness of the algorithms are proven and some important conclusions are obtained. In addition, the proposed algorithms can improve the consistency or consensus of reciprocal [0,1]-valued preference relations with less interactions with the decision makers, which can save a lot of time and obtain the results quickly.
A New Retraction For Accelerating The Riemannian Three-Factor Low-Rank Matrix Completion Algorithm

Zhizhong Li,Deli Zhao,Zhouchen Lin,Edward Y. Chang

DOI: https://doi.org/10.1109/cvpr.2015.7299083

2015-01-01

Abstract:The Riemannian three-factor matrix completion (R3MC) algorithm is one of the state-of-the-art geometric optimization methods for the low-rank matrix completion problem. It is a nonlinear conjugate-gradient method optimizing on a quotient Riemannian manifold. In the line search step, R3MC approximates the minimum point on the searching curve by minimizing on the line tangent to the curve. How-ever, finding the exact minimum point by iteration is too expensive. We address this issue by proposing a new retraction with a minimizing property. This special property provides the exact minimization for the line search by establishing correspondences between points on the searching curve and points on the tangent line. Accelerated R3MC, which is R3MC equipped with this new retraction, outperforms the original algorithm and other geometric algorithms for matrix completion in our empirical study.
Inconsistency indices in pairwise comparisons: an improvement of the Consistency Index

Yuji Sato,Kim Hua Tan

DOI: https://doi.org/10.1007/s10479-021-04431-3

IF: 4.82

2022-01-25

Annals of Operations Research

Abstract:The Consistency Index and the Consistency Ratio of the analytic hierarchy process (AHP) were designed to measure the ratio of inconsistent judgments among pairwise comparisons (PCs), which have been the principal indices for the past four decades. Definitions of inconsistency measures for PCs have yet to be established, however, because of the difficulty in quantifying subjectivity in judgments. Therefore, an empirical review that can take such subjective factors into account is essential. In this paper, the Consistency Ratio is thus reviewed using subjective data, and then a new inconsistency index for PCs is proposed based on the review. The review is based on subjective data obtained from two opinion surveys, which focuses on the relationship between the Consistency Ratio and two indicators: (1) the conformity of the results of the AHP and that of the ranking method, and (2) the goodness-of-fit of weight elicited by the AHP to human perception. A new inconsistency index is then proposed based on the mathematical property of a pairwise comparison matrix and further validated based on the conformity and the goodness-of-fit of weight. The results show that the proposed index detects inconsistency among real-world PCs more sensitively than could the Consistency Ratio; the index might suggest the reliability of the output of a pairwise comparison matrix.

operations research & management science
Compatible Matrices of Spearman's Rank Correlation

Bin Wang,Ruodu Wang,Yuming Wang

DOI: https://doi.org/10.48550/arXiv.1810.03477

2019-04-04

Abstract:In this paper, we provide a negative answer to a long-standing open problem on the compatibility of Spearman's rho matrices. Following an equivalence of Spearman's rho matrices and linear correlation matrices for dimensions up to 9 in the literature, we show non-equivalence for dimensions 12 or higher. In particular, we connect this problem with the existence of a random vector under some linear projection restrictions in two characterization results.

Statistics Theory,Applications
Deriving priorities from pairwise comparison matrices with a novel consistency index

Fang Liu,Shu-Cai Zou,Qing Li

DOI: https://doi.org/10.1016/j.amc.2020.125059

IF: 4.397

2020-06-01

Applied Mathematics and Computation

Abstract:It is important to measure the inconsistency level of a pairwise comparison matrix (PCM) and derive the priority vector in the analytic hierarchy process (AHP). In the present study, a new consistency index is proposed by using the cosine similarity measures of two row/column vectors in a PCM. It is called as the double cosine similarity consistency index (DCSCI) since the row and column vectors are all considered. Some interesting properties of DCSCI are investigated and the thresholds for inconsistency tolerance level are discussed in detail. Then following the idea of DCSCI, we provide a new method for obtaining the priority vector from a PCM. Through maximizing the sum of the cosine similarity measures between the priority vector and the row/column vectors, the priority vector is derived by solving the constructed optimization problem. By analyzing the proposed double cosine similarity maximization (DCSM) method, it is found that the existing cosine maximization (CM) method can be retrieved. By carrying out numerical examples, some comparisons with the existing methods show that the proposed index and method are effective and flexible.

mathematics, applied

A New Index to Measure the Consistency of Reciprocal Matrix

Geometric consistency index for interval pairwise comparison matrices

CONVERGENCE AND STABILITY OF ITERATIVELY REWEIGHTED LEAST SQUARES FOR LOW-RANK MATRIXRECOVERY

Further Discussion on C.R.

Geometric mean method for judgement matrices: Formulas for errors

Consistency and Consensus Improving Methods for Pairwise Comparison Matrices Based on Abelian Linearly Ordered Group

Efficiency analysis for the Perron vector of a reciprocal matrix

New inconsistency indicators for incomplete pairwise comparisons matrices

Efficiency of the convex hull of the columns of certain triple perturbed consistent matrices

Efficient Vectors for Block Perturbed Consistent Matrices

Optimizing consistency and consensus in group decision making based on relative projection between multiplicative reciprocal matrices

Analysis of natural cardinal ranking vectors for pairwise comparisons and the universal efficiency of the Perron geometric mean

A gradient method for inconsistency reduction of pairwise comparisons matrices

A New Method of Improving the Consistency of Fuzzy Judgement Matrices Based on Compatibility

A VARIANT OF THE REFINED ARNOLDI METHOD FOR LARGE SCALE MATRIX EIGENPROBLEMS

Algorithms for improving consistency or consensus of reciprocal [0,1]-valued preference relations

A New Retraction For Accelerating The Riemannian Three-Factor Low-Rank Matrix Completion Algorithm

Inconsistency indices in pairwise comparisons: an improvement of the Consistency Index

Compatible Matrices of Spearman's Rank Correlation

Deriving priorities from pairwise comparison matrices with a novel consistency index