arXiv subjects

Yuzo Maruyama

Publications and source records attributed to Yuzo Maruyama.

30 records · Page 2Linked to original sources

arXiv2015

An alternative to Moran's I for spatial autocorrelation

Moran's I statistic, a popular measure of spatial autocorrelation, is revisited. The exact range of Moran's I is given as a function of spatial weights matrix. We demonstrate that some spatial weights matrices lead the absolute value of upper (lower) bound larger than 1 and that others lead the lower bound larger than -0.5. Thus Moran's I is unlike Pearson's correlation coefficient. It is also pointed out that some spatial weights matrices do not allow Moran's I to take positive values regardless of observations. An alternative measure with exact range [-1,1] is proposed through a monotone transformation of Moran's I.

math.ST↗

arXiv2014

Inadmissibility of the best equivariant predictive density in the unknown variance case

In this work, we are concerned with the estimation of the predictive density of a Gaussian random vector where both the mean and the variance are unknown. In such a context, we prove the inadmissibility of the best equivariant predictive density under the Kullback-Leibler risk in a nonasymptotic framework. Our result stands whatever the dimension d of the vector is, even when d<=2, which can be somewhat surprising compared to the known variance setting. We also propose a class of priors leading to a Bayesian predictive density that dominates the best equivariant one. Throughout the article, we give several elements that we believe are useful for establishing the parallel between the prediction and the estimation problems, as it was done in the known variance framework.

math.ST↗

arXiv2013

Improved robust Bayes estimators of the error variance in linear models

We consider the problem of estimating the error variance in a general linear model when the error distribution is assumed to be spherically symmetric, but not necessary Gaussian. In particular we study the case of a scale mixture of Gaussians including the particularly important case of the multivariate-t distribution. Under Stein's loss, we construct a class of estimators that improve on the usual best unbiased (and best equivariant) estimator. Our class has the interesting double robustness property of being simultaneously generalized Bayes (for the same generalized prior) and minimax over the entire class of scale mixture of Gaussian distributions.

math.ST↗

arXiv2013

Robust Bayesian variable selection with sub-harmonic priors

This paper studies Bayesian variable selection in linear models with general spherically symmetric error distributions. We propose sub-harmonic priors which arise as a class of mixtures of Zellner's g-priors for which the Bayes factors are independent of the underlying error distribution, as long as it is in the spherically symmetric class. Because of this invariance to spherically symmetric error distribution, we refer to our method as a robust Bayesian variable selection method. We demonstrate that our Bayes factors have model selection consistency and are coherent. We also develop Laplace approximations to Bayes factors for a number of recently studied mixtures of g-priors that have recently appeared in the literature (including our own) for Gaussian errors. These approximations, in each case, are given by the Gaussian Bayes factor based on BIC times a simple rational function of the prior's hyper-parameters and the R^2's for the respective models. We also extend model selection consistency for several g-prior based Bayes factor methods for Gaussian errors to the entire class of spherically symmetric error distributions. Additionally we demonstrate that our class of sub-harmonic priors are the only ones within a large class of mixtures of g-priors studied in the literature which are robust in our sense. A simulation study and an analysis of two real data sets indicates good performance of our robust Bayes factors relative to BIC and to other mixture of g-prior based methods.

stat.ME↗

arXiv2012

A Bayes factor with reasonable model selection consistency for ANOVA model

For the ANOVA model, we propose a new g-prior based Bayes factor without integral representation, with reasonable model selection consistency for any asymptotic situations (either number of levels of the factor and/or number of replication in each level goes to infinity). Exact analytic calculation of the marginal density under a special choice of the priors enables such a Bayes factor.

stat.ME↗

arXiv2012

A new Monte Carlo sampling in Bayesian probit regression

We study probit regression from a Bayesian perspective and give an alternative form for the posterior distribution when the prior distribution for the regression parameters is the uniform distribution. This new form allows simple Monte Carlo simulation of the posterior as opposed to MCMC simulation studied in much of the literature and may therefore be more efficient computationally. We also provide alternative explicit expression for the first and second moments. Additionally we provide analogous results for Gaussian priors.

stat.ME↗

arXiv2012

Fully Bayes factors with a generalized g-prior

For the normal linear model variable selection problem, we propose selection criteria based on a fully Bayes formulation with a generalization of Zellner's $g$-prior which allows for $p>n$. A special case of the prior formulation is seen to yield tractable closed forms for marginal densities and Bayes factors which reveal new model evaluation characteristics of potential interest.

stat.ME↗

arXiv2010

Bayesian predictive densities for linear regression models under alpha-divergence loss: some results and open problems

This paper considers estimation of the predictive density for a normal linear model with unknown variance under alpha-divergence loss for -1 <= alpha <= 1. We first give a general canonical form for the problem, and then give general expressions for the generalized Bayes solution under the above loss for each alpha. For a particular class of hierarchical generalized priors studied in Maruyama and Strawderman (2005, 2006) for the problems of estimating the mean vector and the variance respectively, we give the generalized Bayes predictive density. Additionally, we show that, for a subclass of these priors, the resulting estimator dominates the generalized Bayes estimator with respect to the right invariant prior when alpha=1, i.e., the best (fully) equivariant minimax estimator.

math.ST↗

arXiv2008

An extended class of minimax generalized Bayes estimators of regression coefficients

We derive minimax generalized Bayes estimators of regression coefficients in the general linear model with spherically symmetric errors under invariant quadratic loss for the case of unknown scale. The class of estimators generalizes the class considered in Maruyama and Strawderman (2005) to include non-monotone shrinkage functions.

math.ST↗

arXiv2007

Some notes on improving upon the James-Stein estimator

We consider estimation of a multivariate normal mean vector under sum of squared error loss. We propose a new class of smooth estimators parameterized by αdominating the James-Stein estimator. The estimator for α=1 corresponds to the generalized Bayes estimator with respect to the harmonic prior. When αgoes to infinity, the estimator converges to the James-Stein positive-part estimator. Thus the class of our estimators is a bridge between the admissible estimator (α=1) and the inadmissible estimator (α=\infty). Although the estimators have quasi-admissibility which is a weaker optimality than admissibility, the problem of determining whether or not the estimator for α>1 admissible is still open.

math.ST↗

arXiv2006

Admissibility and minimaxity of generalized Bayes estimators for spherically symmetric family

We give a sufficient condition for admissibility of generalized Bayes estimators of the location vector of spherically symmetric distribution under squared error loss. Compared to the known results for the multivariate normal case, our sufficient condition is very tight and is close to being a necessary condition. In particular we establish the admissibility of generalized Bayes estimators with respect to the harmonic prior and priors with slightly heavier tail than the harmonic prior. We use the theory of regularly varying functions to construct a sequence of smooth proper priors approaching an improper prior fast enough for establishing the admissibility. We also discuss conditions of minimaxity of the generalized Bayes estimator with respect to the harmonic prior.

math.ST↗

arXiv2005

A new class of generalized Bayes minimax ridge regression estimators

Let y=Aβ+ε, where y is an N\times1 vector of observations, βis a p\times1 vector of unknown regression coefficients, A is an N\times p design matrix and εis a spherically symmetric error term with unknown scale parameter σ. We consider estimation of βunder general quadratic loss functions, and, in particular, extend the work of Strawderman [J. Amer. Statist. Assoc. 73 (1978) 623-627] and Casella [Ann. Statist. 8 (1980) 1036-1056, J. Amer. Statist. Assoc. 80 (1985) 753-758] by finding adaptive minimax estimators (which are, under the normality assumption, also generalized Bayes) of β, which have greater numerical stability (i.e., smaller condition number) than the usual least squares estimator. In particular, we give a subclass of such estimators which, surprisingly, has a very simple form. We also show that under certain conditions the generalized Bayes minimax estimators in the normal case are also generalized Bayes and minimax in the general case of spherically symmetric errors.

math.ST↗