Simultaneous confidence bands for restricted logistic regression models

Lucy Kerns and John T. Chen

Journal of Applied Statistics, 2017, vol. 44, issue 11, 2036-2051

Abstract: The hyperbolic $ 1-\alpha $ 1−α confidence bands for one logistic regression model with restricted predictors have been considered in the statistical literature. At times, one wishes to construct simultaneous confidence bands for comparing several logistic regression models. It seems that Liu's book [Simultaneous Inference in Regression, Chapman & Hall, 2010, Chapter 8] is the only published work that has addressed this problem. Liu suggested simulation-based methods for constructing simultaneous confidence bands for comparing several logistic models, but further research was warranted to assess the conservativeness of the bands. In this paper, we propose a dimension-wise partitioning method to construct a set of simultaneous confidence bands for the comparisons of several logistic regression functions with a pre-specified function in a stepwise fashion. In addition, simulation studies cast new light on the assumption of predetermined testing order for the stepwise procedures presented in this paper and by Hsu and Berger [Stepwise confidence intervals without multiplicity adjustment for dose–response and toxicity studies, J. Amer. Statist. Assoc. 94 (1999), pp. 468–482]. As an illustration, we include an example on the success rate of thrombolysis associated with patient characteristics regarding post-thrombotic syndrome.

Date: 2017
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/02664763.2016.1238056 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:japsta:v:44:y:2017:i:11:p:2036-2051

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/CJAS20

DOI: 10.1080/02664763.2016.1238056

Access Statistics for this article

Journal of Applied Statistics is currently edited by Robert Aykroyd

More articles in Journal of Applied Statistics from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().