@ARTICLE{TreeBASE2Ref16254,
author = {Nicolas Lartillot and Herv? Philippe},
title = {Computing Bayes Factors Using Thermodynamic Integration.},
year = {2006},
keywords = {},
doi = {10.1080/10635150500433722},
url = {},
pmid = {},
journal = {Systematic Biology},
volume = {55},
number = {2},
pages = {195--207},
abstract = {In the Bayesian paradigm, a common method for comparing two models is to compute the Bayes factor, defined as the ratio of their respective marginal likelihoods. In recent phylogenetic works, the numerical evaluation of marginal likelihoods has often been performed using the harmonic mean estimation procedure. In the present paper, we propose to employ another method, based on an analogy with statistical physics, called thermodynamic integration. We describe the method, propose an implementation, and show on two analytical examples that this numerical method yields reliable estimates. In contrast, the harmonic mean estimator leads to a strong over-estimation of the marginal likelihood, which is all the more pronounced as the model is higher-dimensional. As a result, the harmonic mean estimator systematically favors more parameter-rich models, an artefact which might explain some recent puzzling observations, based on harmonic mean estimates, suggesting that Bayes factors tend to overscore complex models. Finally, we apply our method to the comparison of several alternative models of amino-acid replacement. We confirm our previous observations, indicating that modeling pattern heterogeneity across sites tends to yield better models than standard empirical matrices.}
}
Citation for Study 1452
Citation title:
"Computing Bayes Factors Using Thermodynamic Integration.".
This study was previously identified under the legacy study ID S1388
(Status: Published).
Citation
Lartillot N., & Philippe H. 2006. Computing Bayes Factors Using Thermodynamic Integration. Systematic Biology, 55(2): 195-207.
Authors
Abstract
In the Bayesian paradigm, a common method for comparing two models is to compute the Bayes factor, defined as the ratio of their respective marginal likelihoods. In recent phylogenetic works, the numerical evaluation of marginal likelihoods has often been performed using the harmonic mean estimation procedure. In the present paper, we propose to employ another method, based on an analogy with statistical physics, called thermodynamic integration. We describe the method, propose an implementation, and show on two analytical examples that this numerical method yields reliable estimates. In contrast, the harmonic mean estimator leads to a strong over-estimation of the marginal likelihood, which is all the more pronounced as the model is higher-dimensional. As a result, the harmonic mean estimator systematically favors more parameter-rich models, an artefact which might explain some recent puzzling observations, based on harmonic mean estimates, suggesting that Bayes factors tend to overscore complex models. Finally, we apply our method to the comparison of several alternative models of amino-acid replacement. We confirm our previous observations, indicating that modeling pattern heterogeneity across sites tends to yield better models than standard empirical matrices.
External links
About this resource
- Canonical resource URI:
http://purl.org/phylo/treebase/phylows/study/TB2:S1452
- Other versions:
Nexus
NeXML
- Show BibTeX reference
@ARTICLE{TreeBASE2Ref16254,
author = {Nicolas Lartillot and Herv? Philippe},
title = {Computing Bayes Factors Using Thermodynamic Integration.},
year = {2006},
keywords = {},
doi = {10.1080/10635150500433722},
url = {},
pmid = {},
journal = {Systematic Biology},
volume = {55},
number = {2},
pages = {195--207},
abstract = {In the Bayesian paradigm, a common method for comparing two models is to compute the Bayes factor, defined as the ratio of their respective marginal likelihoods. In recent phylogenetic works, the numerical evaluation of marginal likelihoods has often been performed using the harmonic mean estimation procedure. In the present paper, we propose to employ another method, based on an analogy with statistical physics, called thermodynamic integration. We describe the method, propose an implementation, and show on two analytical examples that this numerical method yields reliable estimates. In contrast, the harmonic mean estimator leads to a strong over-estimation of the marginal likelihood, which is all the more pronounced as the model is higher-dimensional. As a result, the harmonic mean estimator systematically favors more parameter-rich models, an artefact which might explain some recent puzzling observations, based on harmonic mean estimates, suggesting that Bayes factors tend to overscore complex models. Finally, we apply our method to the comparison of several alternative models of amino-acid replacement. We confirm our previous observations, indicating that modeling pattern heterogeneity across sites tends to yield better models than standard empirical matrices.}
}
- Show RIS reference
TY - JOUR
ID - 16254
AU - Lartillot,Nicolas
AU - Philippe,Herv?
T1 - Computing Bayes Factors Using Thermodynamic Integration.
PY - 2006
KW -
UR - http://dx.doi.org/10.1080/10635150500433722
N2 - In the Bayesian paradigm, a common method for comparing two models is to compute the Bayes factor, defined as the ratio of their respective marginal likelihoods. In recent phylogenetic works, the numerical evaluation of marginal likelihoods has often been performed using the harmonic mean estimation procedure. In the present paper, we propose to employ another method, based on an analogy with statistical physics, called thermodynamic integration. We describe the method, propose an implementation, and show on two analytical examples that this numerical method yields reliable estimates. In contrast, the harmonic mean estimator leads to a strong over-estimation of the marginal likelihood, which is all the more pronounced as the model is higher-dimensional. As a result, the harmonic mean estimator systematically favors more parameter-rich models, an artefact which might explain some recent puzzling observations, based on harmonic mean estimates, suggesting that Bayes factors tend to overscore complex models. Finally, we apply our method to the comparison of several alternative models of amino-acid replacement. We confirm our previous observations, indicating that modeling pattern heterogeneity across sites tends to yield better models than standard empirical matrices.
L3 - 10.1080/10635150500433722
JF - Systematic Biology
VL - 55
IS - 2
SP - 195
EP - 207
ER -