Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

PARTIAL MANTEL TESTS: REPLY TO CASTELLANO AND BALLETTO

PARTIAL MANTEL TESTS: REPLY TO CASTELLANO AND BALLETTO PARTIAL MANTEL TESTS: REPLY TO CASTELLANO AND BALLETTO FRANCOIS ROUSSET ¸ CC065, Institut des Sciences de l’Evolution, Universite des Sciences et Techniques du Languedoc, Pl. E. Bataillon, ´ 34095 Montpellier Cedex, France Email: rousset@isem.univ-montp2.fr Received June 26, 2002. Accepted June 26, 2002. In our note (Raufaste and Rousset 2001), we gave an example where the permutation procedure of the partial Mantel test is invalid. From this, we concluded that partial Mantel tests are generally not valid. Castellano and Balletto (2002) report that they find that the partial Mantel test has correct type-1 error probabilities in ‘‘a more general model of causal relationship’’ than the one we considered. Why the difference? For the purpose of providing an example, we deliberately chose simplifying assumptions (populations set around a circle and migration between adjacent populations only; Raufaste and Rousset 2001). Castellano and Balletto (2002) object to these assumptions, but provide no evidence that these particular assumptions are responsible for our results. Of course, a practically reliable testing procedure should not depend on the exact position of samples. Further, localized dispersal between nonadjacent populations could have been incorporated in our analyses without altering our conclusions. Castellano and Balletto’s (2002) results, however, are http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Evolution Wiley

PARTIAL MANTEL TESTS: REPLY TO CASTELLANO AND BALLETTO

Rousset, Francois
Evolution , Volume 56 (9) – Sep 1, 2002
Download PDF
2 pages

Loading next page...
 
/lp/wiley/partial-mantel-tests-reply-to-castellano-and-balletto-5TartpTNpO

References (11)

Publisher
Wiley
Copyright
Copyright © 2002 Wiley Subscription Services, Inc., A Wiley Company
ISSN
0014-3820
eISSN
1558-5646
DOI
10.1111/j.0014-3820.2002.tb00204.x
Publisher site
See Article on Publisher Site

Abstract

PARTIAL MANTEL TESTS: REPLY TO CASTELLANO AND BALLETTO FRANCOIS ROUSSET ¸ CC065, Institut des Sciences de l’Evolution, Universite des Sciences et Techniques du Languedoc, Pl. E. Bataillon, ´ 34095 Montpellier Cedex, France Email: rousset@isem.univ-montp2.fr Received June 26, 2002. Accepted June 26, 2002. In our note (Raufaste and Rousset 2001), we gave an example where the permutation procedure of the partial Mantel test is invalid. From this, we concluded that partial Mantel tests are generally not valid. Castellano and Balletto (2002) report that they find that the partial Mantel test has correct type-1 error probabilities in ‘‘a more general model of causal relationship’’ than the one we considered. Why the difference? For the purpose of providing an example, we deliberately chose simplifying assumptions (populations set around a circle and migration between adjacent populations only; Raufaste and Rousset 2001). Castellano and Balletto (2002) object to these assumptions, but provide no evidence that these particular assumptions are responsible for our results. Of course, a practically reliable testing procedure should not depend on the exact position of samples. Further, localized dispersal between nonadjacent populations could have been incorporated in our analyses without altering our conclusions. Castellano and Balletto’s (2002) results, however, are

Journal

Evolution – Wiley

Published: Sep 1, 2002

There are no references for this article.