STATISTICAL ANALYSIS OF THE m RANKINGS PROBLEM: KENDALL'S PROTOTYPE AND THE GROUPING ESTIMATE METHOD BASED ON NORMAL SCORES

PDF

Authors
  1. Vu, T.
Corporate Authors
Operational Research and Analysis Establishment, Ottawa ONT (CAN) Directorate of Mathematics and Statistics
Abstract
Historically, statistical analysis of the theory of ranks has been developed extensively over the past fifty years. This subject considers the relationship between the rankings that a group of m judges assigns to a set of n objects. Kendall's contributions to the m rankings problem are fundamental ideas for this paper. They include the formulae of the correlation coefficient between two rankings and the coefficient of Concordance W among judges, significance tests of W as well as the "best" estimate based on the order of the sums of natural number ranks for each object. Based on the concepts of normal scores and order statistics, a new procedure to find the "best" estimate is developed in which groups of objects which should be ranked the same are formed based on the test of consecutive order statistics of normal score ranks' sums for each object after sorting in ascending order.
Report Number
ORAE-DMS-RN-4/92 — Research Note
Date of publication
15 Aug 1992
Number of Pages
94
DSTKIM No
93-00139
CANDIS No
127380
Format(s):
Hardcopy;Originator's fiche received by DSIS;Document Image stored on Optical Disk

Permanent link

Document 1 of 1

Date modified: