Following a Bayesian statistical inference paradigm, we provide an alternative methodology for analyzing a multivariate logistic regression. We use a multivariate normal prior in the Bayesian analysis. We present a unique Bayes estimator associated with a prior which is admissible. The Bayes estimators of the coefficients of the model are obtained via MCMC methods. The proposed procedure...

We study the Bayesian approach to nonparametric function estimation problems such as nonparametric regression and signal estimation. We consider the asymptotic properties of Bayes procedures for conjugate (=Gaussian) priors. We show that so long as the prior puts nonzero measure on the very large parameter set of interest then the Bayes estimators are not satisfactory. More specifically, we sho...

An estimation problem of the mean µ of an inverse Gaussian distribution IG(µ, C µ) with known coefficient of variation c is treated as a decision problem with entropy loss function. A class of Bayes estimators is constructed, and shown to include MRSE estimator as its closure. Two important members of this class can easily be computed using continued fractions

There is still lack of clarity about the best manner in which to handle numeric attributes when applying Bayesian network classifiers. Discretization methods entail an unavoidable loss of information. Nonetheless, a number of studies have shown that appropriate discretization can outperform straightforward use of common, but often unrealistic parametric distribution (e.g. Gaussian). Previous st...

The order-q Tsallis (Hq ) and Rényi entropy (Kq ) receive broad applications in the statistical analysis of complex phenomena. A generic problem arises, however, when these entropies need to be estimated from observed data. The finite size of data sets can lead to serious systematic and statistical errors in numerical estimates. In this paper, we focus upon the problem of estimating generalized...