betaFeedback

WorksMathematics

Worst-case bias of the independence assumption for functions of two ordinal variables

Abstract

Let ordinal grades XX and YY be obtained by cutting a pair of random variables with copula CC at arbitrary thresholds, and let gg be a function of the grades. We study the largest error SC(g)S_C(g), over all thresholds, that is, over all marginal laws, made when the mean of g(X,Y)g(X,Y) is computed as if the grades were independent. This is the counterpart of the classical dependence-uncertainty problem: the copula is fixed and the discretisation is free. The error is caused entirely by the interaction of the inputs, measured by an index ρ(g)ρ(g), the largest absolute gg-volume of a rectangle of grades, which vanishes if and only if gg is additive. For every copula, SC(g)S_C(g) lies between ρ(g)ρ(g) times the largest deviation of CC from the independence copula and ρ(g)ρ(g) times half the total variation of their difference, and the factor one half cannot be improved. For the Gaussian copula with correlation λλ we prove that SC(g)S_C(g) equals ∣λ∣ρ(g)/(2π)|λ|ρ(g)/(2π) to first order, with a sharp constant and an explicit remainder. The same first-order law holds for copula families whose first-order departure from independence is a product of unimodal factors, including the Farlie-Gumbel-Morgenstern, Ali-Mikhail-Haq, Frank, Clayton and Plackett families, and we characterise the Archimedean families with this property. An application to the TNM staging of colon cancer, with data from a clinical trial, compares the bounds with the actual error of the independence computation.

Paper details

vlnc:W_wd95kryc
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 3

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Claude

View statement
Looking for related papers…