Alternative rank-based evaluations are based on NDCG, and Kendall’s Tau, and Spearman’s Rank Correlation Coeﬃcient .
A Comparative Study of Collaborative Filtering Algorithms
Another recent approach has been based on the calculation of Kendall’s rank τ correlation coefﬁcient between the predicted and actual values, so disregarding the magnitudes.
A graphical analysis of cost-sensitive regression problems
Kendall correlation between two rank orders (typically between the true order and the method’s result) as a measure of performance.
Random Utility Theory for Social Choice
For example, the ranking obtained by using half of the data can still achieve a fair estimate to the results with full data, with an average Kendall correlation of greater than 0.4.
Random Utility Theory for Social Choice
Kendall, Rank correlation methods, Griﬃn, London, 1970.
On asymptotic efficiency of multivariate version of Spearman's rho
Kendall, M. G. (1938). A new measure of rank correlation.
A consistent test of independence based on a sign covariance related to Kendall's tau
Rank-based correlation test like Spearman’s or Kendall’s correlation coeﬃcient are examples of tests that fulﬁll assumption (B2).
Asymptotic optimality of the Westfall--Young permutation procedure for multiple testing under dependence
In the (cid:12)rst subsection we give a brief introduction and background references for Kendall's rank correlation coe(cid:14)cient and Kendall's partial rank correlation coe(cid:14)cient with uncensored data.
A test for partial correlation with censored astronomical data
In spite of the long history of Kendall's partial rank correlation coe(cid:14)cient, there are no tests for the signi(cid:12)cance of the partial (cid:28) (Hettmansperger ).
A test for partial correlation with censored astronomical data
About the rank correlation and their comparison, the literature is quite large, of the recent publications we may cite and where the authors introduce a new measure starting from the the Kendall’s coeﬃcient for the information retrieval ﬁeld.
On the Weakenesses of Correlation Measures used for Search Engines' Results (Unsupervised Comparison of Search Engine Rankings)
As an usual test very close to the linear rank test, we cite the Kendall rank correlation coeﬃcient τn (see ).
Bahadur efficiency of nonparametric test for independence based on $L_1$-error
***