Home > Publications . Search All . Browse All . Country . Browse PSC Pubs . PSC Report Series

PSC In The News

RSS Feed icon

Thompson says America must "unchoose" policies that have led to mass incarceration

Axinn says new data on campus rape will "allow students to see for themselves the full extent of this problem"

Frey says white population is growing in Detroit and other large cities


Susan Murphy to speak at U-M kickoff for data science initiative, Oct 6, Rackham

Andrew Goodman-Bacon, former trainee, wins 2015 Nevins Prize for best dissertation in economic history

Deirdre Bloome wins ASA award for work on racial inequality and intergenerational transmission

Bob Willis awarded 2015 Jacob Mincer Award for Lifetime Contributions to the Field of Labor Economics

Next Brown Bag

Monday, Oct 12 at noon, 6050 ISR
Joe Grengs: Policy & planning for transportation equity

Adaptive Confidence Intervals for the Test Error in Classification

Publication Abstract

Laber, Eric B., and Susan A. Murphy. 2011. "Adaptive Confidence Intervals for the Test Error in Classification." Journal of the American Statistical Association, 106(495): 904-913.

The estimated test error of a learned classifier is the most commonly reported measure of classifier performance. However, constructing a high-quality point estimator of the test error has proved to be very difficult. Furthermore, common interval estimators (e.g., confidence intervals) are based on the point estimator of the test error and thus inherit all the difficulties associated with the point estimation problem. As a result, these confidence intervals do not reliably deliver nominal coverage. In contrast, we directly construct the confidence interval by using smooth data-dependent upper and lower bounds on the test error. We prove that, for linear classifiers, the proposed confidence interval automatically adapts to the nonsmoothness of the test error, is consistent under fixed and local alternatives, and does not require that the Bayes classifier be linear. Moreover, the method provides nominal coverage on a suite of test problems using a range of classification algorithms and sample sizes. This article has supplementary material online.

DOI:10.1198/jasa.2010.tm10053 (Full Text)

PMCID: PMC3285493. (Pub Med Central)

Browse | Search : All Pubs | Next