Andrea Mignatti, Giorgio Corani

---

Credal model averaging of logistic regression for modeling the distribution of marmot burrows

Abstract

Credal model averaging (CMA) is a credal ensemble of Bayesian models, which generalizes Bayesian model averaging (BMA). An open problem of BMA is how to set the prior over the models. CMA overcomes this problem by substituting the single prior over the models by a set of priors. We devise CMA for logistic regression; the different logistic regressors, over which the credal averaging is performed, are characterized by different feature sets. CMA returns indeterminate classifications when the classification is prior-dependent, namely when the most probable class (presence or absence) depends on the prior which is set over the models. We apply CMA for modelling the distribution of marmot burrows in an Alpine valley in Italy.

Keywords

Bayesian model averaging, credal model averaging, logistic regression, classification, ecological modelling

---

Download area

The paper is available in the following formats:

• pdf file

E-mail addresses

Andrea Mignatti		mignatti@elet.polimi.it
Giorgio Corani		giorgio@idsia.ch