A prior near-ignorance Gaussian process model for nonparametric regression

This paper proposes a prior near-ignorance model for regression based on a set of Gaussian Processes (GP). GPs are natural prior distributions for Bayesian regression. They offer a great modeling flexibility and have found widespread application in many regression problems. However, a GP requires the prior elicitation of its mean function, which represents our prior belief about the shape of the regression function, and of the covariance between any two function values.In the absence of prior information, it may be difficult to fully specify these infinite dimensional parameters. In this work, by modeling the prior mean of the GP as a linear combination of a set of basis functions and assuming as prior for the combination coefficients a set of conjugate distributions obtained as limits of truncate exponential priors, we have been able to model prior ignorance about the mean of the GP. The resulting model satisfies translation invariance, learning and, under some constraints, convergence, which are desirable properties for a prior near-ignorance model. Moreover, it is shown in this paper how this model can be extended to allow for a weaker specification of the GP covariance between function values, by letting each basis function to vary in a set of functions.Application to hypothesis testing has shown how the use of this model induces the capability of automatically detecting when a reliable decision cannot be made based on the available data.     

Wavelet modeling of priors on triangles

Parameters in statistical problems often live in a geometry of certain shape. For example, count probabilities in a multinomial distribution belong to a simplex. For these problems, Bayesian analysis needs to model priors satisfying certain constraints imposed by the geometry. This paper investigates modeling of priors on triangles by use of wavelets constructed specifically for triangles. Theoretical analysis and numerical simulations show that our modeling is flexible and is superior to the commonly used Dirichlet prior.     

Skew normal measurement error models

In this paper we define a class of skew normal measurement error models, extending usual symmetric normal models in order to avoid data transformation. The likelihood function of the observed data is obtained, which can be maximized by using existing statistical software. Inference on the parameters of interest can be approached by using the observed information matrix, which can also be computed by using existing statistical software, such as the Ox program. Bayesian inference is also discussed for the family of asymmetric models in terms of invariance with respect to the symmetric normal distribution showing that early results obtained for the normal distribution also holds for the asymmetric family. Results of a simulation study and an analysis of a real data set analysis are provided.     

正态分布在不同先验下的Bayes估计及风险比较

本文讨论在均值未知,方差已知的正态分布情况下通过在共轭先验以及Jeffreys先验二种先验下的Bayes估计问题,在平方损失函数下和线性损失函数下Bayes风险的比较.数据计算可以看出,在Jeffreys先验下的Bayes风险要比在共轭先验下的Bayes风险要大,但是当样本量增大时,两者的后验风险越来越靠近.     

An ensemble method using credal decision trees

Supervised classification learning can be considered as an important tool for decision support. In this paper, we present a method for supervised classification learning, which ensembles decision trees obtained via convex sets of probability distributions (also called credal sets) and uncertainty measures. Our method forces the use of different decision trees and it has mainly the following characteristics: it obtains a good percentage of correct classifications and an improvement in time of processing compared with known classification methods; it not needs to fix the number of decision trees to be used; and it can be parallelized to apply it on very large data sets.     

A note on the representation of parametric models for multivariate extremes

In this note, the representations of extremal Dirichlet and logistic distributions are reviewed and extended. These new representations allow exact simulations of the spectral distribution functions and an extension of the extremal logistic case to dimensions higher than two.      

Credal model averaging for classification: representing prior ignorance and expert opinions

Bayesian model averaging (BMA) is the state of the art approach for overcoming model uncertainty. Yet, especially on small data sets, the results yielded by BMA might be sensitive to the prior over the models. Credal model averaging (CMA) addresses this problem by substituting the single prior over the models by a set of priors (credal set). Such approach solves the problem of how to choose the prior over the models and automates sensitivity analysis. We discuss various CMA algorithms for building an ensemble of logistic regressors characterized by different sets of covariates. We show how CMA can be appropriately tuned to the case in which one is prior-ignorant and to the case in which instead domain knowledge is available. CMA detects prior-dependent instances, namely instances in which a different class is more probable depending on the prior over the models. On such instances CMA suspends the judgment, returning multiple classes. We thoroughly compare different BMA and CMA variants on a real case study, predicting presence of Alpine marmot burrows in an Alpine valley. We find that BMA is almost a random guesser on the instances recognized as prior-dependent by CMA.     

A tree augmented classifier based on Extreme Imprecise Dirichlet Model

We present TANC, a TAN classifier (tree-augmented naive) based on imprecise probabilities. TANC models prior near-ignorance via the Extreme Imprecise Dirichlet Model (EDM). A first contribution of this paper is the experimental comparison between EDM and the global Imprecise Dirichlet Model using the naive credal classifier (NCC), with the aim of showing that EDM is a sensible approximation of the global IDM. TANC is able to deal with missing data in a conservative manner by considering all possible completions (without assuming them to be missing-at-random), but avoiding an exponential increase of the computational time. By experiments on real data sets, we show that TANC is more reliable than the Bayesian TAN and that it provides better performance compared to previous TANs based on imprecise probabilities. Yet, TANC is sometimes outperformed by NCC because the learned TAN structures are too complex; this calls for novel algorithms for learning the TAN structures, better suited for an imprecise probability classifier.     

单纯形分布非线性模型的局部影响分析及其应用

讨论了单纯形分布非线性模型的局部影响分析问题.应用Cook(1986)的影响曲率方法研究了该模型关于微小扰动的局部影响,得到了局部影响分析的曲率度量.同时也应用PoonW Y和Poon Y S(1997)的保形法曲率方法研究了该模型的局部影响.对常见的扰动模型,分别进行了局部影响分析,得到了计算影响矩阵的简洁公式.最后还研究了两个实例,说明文中方法的应用价值.     

服从Dirichlet分布的成分数据的贝叶斯分析

本文研究了Dirichlet分布总体的参数和其他感光趣的量的贝叶斯估计。在参数的有实际意义的函数上设置均匀的先验分布，对适当变换后的参数用Metropolis算法得到马尔可夫链蒙特卡罗后验样本，由此即得参数和其他感兴趣的量的贝叶斯估计。     

Statistical inference with partial prior information based on a Gauss-type inequality

Potter and Anderson [1] have developed a Bayesian decision procedure requiring the specification of a class of prior distributions restricted to have a minimal probability content for a given subset of the parameter space. They do not, however, provide a method for the selection of that subset. We show how a generalization of Gauss' inequality can be used to determine the relevant parameter subset.     

A limit result for the prior predictive applied to checking for prior-data conflict

We consider checking for prior-data conflict in a Bayesian analysis via a tail probability based on the prior predictive distribution. We establish the appropriateness of this measure in the sense that the limiting value of the tail probability measures the extent to which the true value of the parameter is a surprising value from the prior.     

On perturbations of strongly admissible prior distributions

Consider a parametric statistical model, P(dx|θ), and an improper prior distribution, ν(dθ), that together yield a (proper) formal posterior distribution, Q(dθ|x). The prior is called strongly admissible if the generalized Bayes estimator of every bounded function of θ is admissible under squared error loss. Eaton [M.L. Eaton, A statistical diptych: Admissible inferences-recurrence of symmetric Markov chains, Annals of Statistics 20 (1992) 1147–1179] used the Blyth–Stein Lemma to develop a sufficient condition, call it , for strong admissibility of ν. Our main result says that, under mild regularity conditions, if ν satisfies and g(θ) is a bounded, non-negative function, then the perturbed prior distribution g(θ)ν(dθ) also satisfies and is therefore strongly admissible. Our proof has three basic components: (i) Eaton's [M.L. Eaton, A statistical diptych: Admissible inferences-recurrence of symmetric Markov chains, Annals of Statistics 20 (1992) 1147–1179] result that the condition is equivalent to the local recurrence of the Markov chain whose transition function is R(dθ|η)=∫Q(dθ|x)P(dx|η); (ii) a new result for general state space Markov chains giving conditions under which local recurrence is equivalent to recurrence; and (iii) a new generalization of Hobert and Robert's [J.P. Hobert, C.P. Robert, Eaton's Markov chain, its conjugate partner and -admissibility, Annals of Statistics 27 (1999) 361–373] result that says Eaton's Markov chain is recurrent if and only if the chain with transition function is recurrent. One important application of our results involves the construction of strongly admissible prior distributions for estimation problems with restricted parameter spaces.     

A white noise approach to infinitely divisible distributions on Gel'fand triple

In this note, we apply white noise analysis to infinitely divisible distributions on a real Gel'fand triple E⊂H⊂E^∗. We first introduce an index, called Hida index, for a measure on E⊂H⊂E^∗. And then, under some mild conditions, we obtain a general inequality which indicates a connection between the Hida index of an infinitely divisible distribution on E⊂H⊂E^∗ and that of its Lévy measure. Finally we prove that the Hida index of the standard compound Poisson distribution on E⊂H⊂E^∗ is exactly 1.     

On the use of simplex methods in constructing quadratic models

In this paper,we investigate the quadratic approximation methods.After studying the basic idea of simplex methods,we construct several new search directions by combining the local information progressively obtained during the iterates of the algorithm to form new subspaces.And the quadratic model is solved in the new subspaces.The motivation is to use the information disclosed by the former steps to construct more promising directions.For most tested problems,the number of function evaluations have been reduced obviously through our algorithms.     

On the projections of the Dirichlet probability distribution on symmetric cones

In this paper, we introduce the Riesz-Dirichlet distribution on a symmetric cone as an extension of the Dirichlet distribution defined by the Wishart distribution. We also show that some projections of these distributions related to the Pierce decomposition are also Dirichlet.     

Multivariate count data generalized linear models: Three approaches based on the Sarmanov distribution

Starting from the question: What is the accident risk of an insured individual?, we consider that the customer has contracted policies in different insurance lines: motor and home. Three models based on the multivariate Sarmanov distribution are analyzed. Driven by a real data set that takes into account three types of accident risks, two for motor and one for home, three trivariate Sarmanov distributions with generalized linear models (GLMs) for marginals are considered and fitted to the data. To estimate the parameters of these three models, we discuss a method for approaching the maximum likelihood (ML) estimators. Finally, the three models are compared numerically with the simpler trivariate Negative Binomial GLM and with elliptical copula based models.     

Bayesian estimation of the intraclass correlation coefficients in the mixed linear model

The method of determining Bayesian estimators for the special ratios of variance components called the intraclass correlation coefficients is presented. The exact posterior distribution for these ratios of variance components is obtained. The approximate posterior mean of this distribution is also derived. All computations are non-iterative and avoid numerical integration.