Skip Navigation


Systematic Biology Advance Access originally published online on July 1, 2009
Systematic Biology 2009 58(2):211-223; doi:10.1093/sysbio/syp016
This Article
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
58/2/211    most recent
syp016v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Susko, E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Susko, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© Society of Systematic Biologists

Bootstrap Support Is Not First-Order Correct

Edward Susko*

Department of Mathematics and Statistics, Dalhousie University, Halifax, Nova Scotia, Canada B3H 3J5

* Correspondence to be sent to: Department of Mathematics and Statistics, Dalhousie University, Halifix, Nova Scotia, Canada B3H 3J5; E-mail: susko{at}mathstat.dal.ca.


  Abstract

The appropriate interpretation of bootstrap support for splits and the question of what constitutes large bootstrap support have received considerable attention. One desirable interpretation, indeed the interpretation that was put forward when bootstrap support for splits was first introduced, is that 1-minus bootstrap support is a P value for the hypothesis that the split is not well resolved. As a P value, bootstrap support has been argued to be first-order correct. By obtaining the limiting distribution of bootstrap support for a split when maximum likelihood estimation is conducted, it is shown that bootstrap support is not first-order correct and insight is provided into the nature of the problem. Borrowing from earlier results, it is also shown that similar results hold when the neighbor-joining algorithm is used. Examples suggest that bootstrap support is generally conservative as a P value and give insight as to why this is usually the case. The analysis indicates that the problem is largely due to the unusual nature of tree space where boundary trees always have at least 2 neighbors.

Keywords: Bootstrap support; problem of regions; splits; star tree; topology test

Received July 4, 2008; Revised September 22, 2008; Accepted January 14, 2009

Associate Editor: Cécile Ané


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?




Disclaimer: Please note that abstracts for content published before 1996 were created through digital scanning and may therefore not exactly replicate the text of the original print issues. All efforts have been made to ensure accuracy, but the Publisher will not be held responsible for any remaining inaccuracies. If you require any further clarification, please contact our Customer Services Department.