Complete ranking procedures with appropriate loss functions

DOI10.1080/03610929008830456MaRDI QIDQ3978040zbMATH OpenOpenAlexFDO

Authors Marc Sobel

Publication date 25 June 1992

Published in Communications in Statistics: Theory and Methods (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/03610929008830456

zbMATH Keywords

independent normal populationsleast favorable parametric configurationloss functionsmaximum value of the risknumerical simulationrisk performancesample sizevariance ranking

Mathematics Subject Classification ID

Statistical decision theory (62C99) Statistical ranking and selection procedures (62F07)

Review text

The work is an outgrowth and extension of the paper of \textit{J. Reeves} and the author, ibid. 16, 837-854 (1987: Zbl 0614.62025). The authors consider the problem of comparing different evaluations of procedures which rank the variances of k independent normal populations when means are unknown. The procedures are evaluated by the performance of their risk. The sample size, the least favorable parametric configuration and the maximum value of the risk are studied for each procedure. Different loss functions are proposed. When k is small, numerical simulation is used to show which loss functions respond better and which respond worse to increases in sample size. Theoretical results are given for large k.

Recommendations

• Complete ranking procedures with quadratic loss
• Loss function based ranking in two-stage, hierarchical models
• Optimal full ranking from pairwise comparisons
• scientific article; zbMATH DE number 7108569
• Regularized ranking with convex losses and \(\ell^1\)-penalty
• Verification, Model Checking, and Abstract Interpretation
• Simple, Robust and Optimal Ranking from Pairwise Comparisons
• Ranking as function approximation

Cites work

• Complete ranking procedures with quadratic loss
• scientific article; zbMATH DE number 3984308 (Why is no real title available?)
• scientific article; zbMATH DE number 3727393 (Why is no real title available?)
• Optimum Sample Size for Choosing the Population Having the Smallest Variance (Large Sample Size Results)

Cited in

(6)

• Optimal full ranking from pairwise comparisons
• Calibration and regret bounds for order-preserving surrogate losses in learning to rank
• Optimal transitive procedures for comparing large numbers of parameters
• Complete ranking procedures with quadratic loss
• scientific article; zbMATH DE number 7108569 (Why is no real title available?)
• Learning Theory

This page was built for publication: Complete ranking procedures with appropriate loss functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3978040)