We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

stat.ME

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Statistics > Methodology

Title: Confidence intervals of prediction accuracy measures for multivariable prediction models based on the bootstrap-based optimism correction methods

Abstract: In assessing prediction accuracy of multivariable prediction models, optimism corrections are essential for preventing biased results. However, in most published papers of clinical prediction models, the point estimates of the prediction accuracy measures are corrected by adequate bootstrap-based correction methods, but their confidence intervals are not corrected, e.g., the DeLong's confidence interval is usually used for assessing the C-statistic. These naive methods do not adjust for the optimism bias and do not account for statistical variability in the estimation of parameters in the prediction models. Therefore, their coverage probabilities of the true value of the prediction accuracy measure can be seriously below the nominal level (e.g., 95%). In this article, we provide two generic bootstrap methods, namely (1) location-shifted bootstrap confidence intervals and (2) two-stage bootstrap confidence intervals, that can be generally applied to the bootstrap-based optimism correction methods, i.e., the Harrell's bias correction, 0.632, and 0.632+ methods. In addition, they can be widely applied to various methods for prediction model development involving modern shrinkage methods such as the ridge and lasso regressions. Through numerical evaluations by simulations, the proposed confidence intervals showed favourable coverage performances. Besides, the current standard practices based on the optimism-uncorrected methods showed serious undercoverage properties. To avoid erroneous results, the optimism-uncorrected confidence intervals should not be used in practice, and the adjusted methods are recommended instead. We also developed the R package predboot for implementing these methods (this https URL). The effectiveness of the proposed methods are illustrated via applications to the GUSTO-I clinical trial.
Subjects: Methodology (stat.ME); Applications (stat.AP); Computation (stat.CO)
Journal reference: Stat Med. 2021
DOI: 10.1002/sim.9148
Cite as: arXiv:2005.01457 [stat.ME]
  (or arXiv:2005.01457v5 [stat.ME] for this version)

Submission history

From: Hisashi Noma [view email]
[v1] Mon, 4 May 2020 13:15:54 GMT (290kb)
[v2] Tue, 5 May 2020 11:33:59 GMT (289kb)
[v3] Fri, 8 May 2020 08:40:44 GMT (289kb)
[v4] Wed, 30 Jun 2021 10:18:57 GMT (395kb)
[v5] Sun, 25 Jul 2021 12:56:27 GMT (393kb)

Link back to: arXiv, form interface, contact.