ProbXiv
Explorersign in
Problem archiveProblem record

Statement

For a tournament of nn teams, observe jointly independent yijsimBern⁡(θij)y_{ij}\\sim\operatorname{Bern}(\theta_{ij}) for 1≤i≤j≤n1\le i\le j\le n, with yij=1−yjiy_{ij}=1-y_{ji} for j<ij<i, θji=1−θij\theta_{ji}=1-\theta_{ij} for i≠ji\ne j, and θii=0\theta_{ii}=0. Assume there is an unknown ordering of the teams satisfying strong stochastic transitivity: if team jj ranks higher than team kk, then θij≤θik\theta_{ij}\le\theta_{ik} for every team ii. Estimate the probability matrix θ\theta under mean squared error loss. The report claims that the overall least-squares estimator attains the minimax rate O(1/n)O(1/n) up to logarithmic factors, whereas its computationally tractable two-stage estimator has maximum squared-error risk O(n−1/2)O(n^{-1/2}) up to logarithmic factors, with O(1/n)O(1/n) up to logarithmic factors on several specified subclasses.

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.