Affiliations: [a] Beedie School of Business, Simon Fraser University, Burnaby, B.C., Canada. choo@sfu.ca; wedley@sfu.ca | [b] TNO Organisation for Applied Scientific Research, The Hague, The Netherlands
Correspondence:
[*]
Address for correspondence: Beedie School of Business, Simon Fraser University, Burnaby, B. C., Canada, V5A 1S6
Note: [†] Financial support necessary for this project was provided by The Natural Sciences and Engineering Research Council of Canada, Simon Fraser University and TNO Organisation for Applied Scientific Research. The authors thank the anonymous reviewers for valuable suggestions. As well, we thank Mohammed Alqarni and the Editors for assistance in formatting the pre-publication file.
Abstract: A ratio pairwise comparison matrix estimates another matrix of true ratios between objects. From the pairwise comparison matrix, various methods are used to derive a priority vector and associated consistent matrix that also estimates the matrix of true ratios. The distance from the consistent matrix and the true matrix measures the accuracy of a method. The geometric mean is shown to be the only method with error below a basic threshold while being invariant to any reordering and rescaling of columns. Besides being simple to calculate, the geometric mean has excellent performance and many desirable properties.
Keywords: Ratio scale preferences, pairwise ratio comparisons, relevant ratio information
