Eliciting vague but proper maximal entropy priors in Bayesian experiments

Priors elicited according to maximal entropy rules have been used for years in objective and subjective Bayesian analysis. However, when the prior knowledge remains fuzzy or dubious, they often suffer from impropriety which can make them uncomfortable to use. In this article we suggest the formal elicitation of an encompassing family for the standard maximal entropy (ME) priors and the maximal data information (MDI) priors, which can lead to obtain proper families. An interpretation is given in the objective framework of channel coding. In a subjective framework, the performance of the method is shown in a reliability context when flat but proper priors are elicited for the Weibull lifetime distributions. Such priors appear as practical tools for sensitivity studies.     

Comment: A Bayesian Approach to the Nonresponse Problem,Using Covariates à la Freedman

We propose a flexible method to approximate the subjective cumulative distribution function of an economic agent about the future realization of a continuous random variable. The method can closely approximate a wide variety of distributions while maintaining weak assumptions on the shape of distribution functions. We show how moments and quantiles of general functions of the random variable can be computed analytically and/or numerically. We illustrate the method by revisiting the determinants of income expectations in the United States. A Monte Carlo analysis suggests that a quantile-based flexible approach can be used to successfully deal with censoring and possible rounding levels present in the data. Finally, our analysis suggests that the performance of our flexible approach matches that of a correctly specified parametric approach and is clearly better than that of a misspecified parametric approach.     

Analysis of means approach for random factor analysis

Analysis of means (ANOM) is a powerful tool for comparing means and variances in fixed-effects models. The graphical exhibit of ANOM is considered as a great advantage because of its interpretability and its ability to evaluate the practical significance of the mean effects. However, the presence of random factors may be problematic for the ANOM method. In this paper, we propose an ANOM approach that can be applied to test random effects in many different balanced statistical models including fixed-, random- and mixed-effects models. The proposed approach utilizes the range of the treatment averages for identifying the dispersions of the underlying populations. The power performance of the proposed procedure is compared to the analysis of variance (ANOVA) approach in a wide range of situations via a Monte Carlo simulation study. Illustrative examples are used to demonstrate the usefulness of the proposed approach and its graphical exhibits, provide meaningful interpretations, and discuss the statistical and practical significance of factor effects.     

On the analysis of multivariate repeated measures designs

The relationship between the mixed-model analysis and multivariate approach to a repeated measures design with multiple responses is presented. It is shown that by taking the trace of the appropriate submatrix of the hypothesis (error) sums of squares and crossproducts (SSCP) matrix obtained from the multivariate approach, one can get the hypothesis (error) SSCP matrix for the mixed-model analysis. Thus, when analyzing data from a multivariate repeated measures design, it is advantageous to use the multivariate approach because the result of the mixed-model analysis can also be obtained without additional computation.     

An approximation of logarithmic functions in the regression setting

We consider a method of moments approach for dealing with censoring at zero for data expressed in levels when researchers would like to take logarithms. A Box–Cox transformation is employed. We explore this approach in the context of linear regression where both dependent and independent variables are censored. We contrast this method to two others, (1) dropping records of data containing censored values and (2) assuming normality for censored observations and the residuals in the model. Across the methods considered, where researchers are interested primarily in the slope parameter, estimation bias is consistently reduced using the method of moments approach.     

On generalised estimating equations for vector regression

Generalised estimating equations (GEE) for regression problems with vector‐valued responses are examined. When the response vectors are of mixed type (e.g. continuous–binary response pairs), the GEE approach is a semiparametric alternative to full‐likelihood copula methods, and is closely related to Prentice & Zhao's mean‐covariance estimation equations approach. When the response vectors are of the same type (e.g. measurements on left and right eyes), the GEE approach can be viewed as a ‘plug‐in’ to existing methods, such as the vglm function from the state‐of‐the‐art VGAM package in R. In either scenario, the GEE approach offers asymptotically correct inferences on model parameters regardless of whether the working variance–covariance model is correctly or incorrectly specified. The finite‐sample performance of the method is assessed using simulation studies based on a burn injury dataset and a sorbinil eye trial dataset. The method is applied to data analysis examples using the same two datasets, as well as to a trivariate binary dataset on three plant species in the Hunua ranges of Auckland.     

Testable hypotheses in singular fixed linear models

Hypotheses that restrict nonestimable parameters in singular (or overparameterized) fixed linear models are considered nontes table by most aothors and are not allowed by most computer packages. In this article, a different approach is taken and hypotheses are classified as completely testable, partially testable, or nontestable on the basis of the number of degrees of freedom associated with them. The convenience of this approach is illustrated with examples and by developing a related general theory of equivalent hypotheses, reparameterizations, and restrictions. A method of transforming partially testable hypotheses into equivalent completely testable hypotheses is described.     

Editorial collaborators

Traditionally, the bioequivalence of a generic drug with the innovator's product is assessed by comparing their pharmacokinetic profiles determined from the blood or plasma concentration-time curves. This method may only be applicable to formulations where blood drug or metabolites levels adequately characterize absorption and metabolism. For non-systematic drugs categorized by the lack of systemic presence, such as metered dose inhalers (MDI), anti-ulcer agents and topical antifungals and vaginal antifungals, new definition of therapeutic equivalency and criteria for acceptance should be used. When pharmacologic effects of the drugs can be easily measured, pharmacodynamic effect studies can be used to assess the therapeutic equivalence of non-systemic drugs. When analytical methods or other tests cannot be developed to permit use of the pharmacodynamic method, clinical trials to compare one or several clinical endpoints may be the only suitable method to establishing therapeutic equivalence. In this paper we evaluate by Monte-Carlo simulations the fixed sample performances of some two one-sided tests procedures which may be used to assess the therapeutic equivalence of non-systemic drugs with binary clinical endpoints. Formulae of sample size determination for therapeutic equivalence clinical trials are also given.     

Focused estimation for noisy and small data sets: a Bayesian minimum expected loss estimator approach

Central to many inferential situations is the estimation of rational functions of parameters. The mainstream in statistics and econometrics estimates these quantities based on the plug‐in approach without consideration of the main objective of the inferential situation. We propose the Bayesian Minimum Expected Loss (MELO) approach focusing explicitly on the function of interest, and calculating its frequentist variability. Asymptotic properties of the MELO estimator are similar to the plug‐in approach. Nevertheless, simulation exercises show that our proposal is better in situations characterised by small sample sizes and/or noisy data sets. In addition, we observe in the applications that our approach gives lower standard errors than frequently used alternatives when data sets are not very informative.     

A bayesian wls approach to generalized linear models

We use a Bayesian approach to fitting a linear regression model to transformations of the natural parameter for the exponential class of distributions. The usual Bayesian approach is to assume that a linear model exactly describes the relationship among the natural parameters. We assume only that a linear model is approximately in force. We approximate the theta-links by using a linear model obtained by minimizing the posterior expectation of a loss function.While some posterior results can be obtained analytically considerable generality follows from an exact Monte Carlo method for obtaining random samples of parameter values or functions of parameter values from their respective posterior distributions. The approach that is presented is justified for small samples, requires only one-dimensional numerical integrations, and allows for the use of regression matrices with less than full column rank. Two numerical examples are provided.     

Randomized response multivariate designs for categorical data

Three approaches to multivariate estimation for categorical data using randomized response (RR) are described. In the first approach, practical only for 2×2 contingency tables, a multi-proportions design is used. In the second approach, a separate RR trial is used for each variate and it is noted that the multi­variate design matrix of conditional probabilities is given by the Kroneeker product of the univariate design matrices of each trial, provided that the trials are independent of each other in a certain sense. The third approach requires only a single randomization and thus may be viewed as the use of vector response. Finally, a special-purpose bivariate design is presented.     

Penalized minimum-distance estimates in finite mixture models

When finite mixture models are used to fit data, it is sometimes important to estimate the number of mixture components. A nonparametric maximum-likelihood approach may result in too many support points and, in general, does not yield a consistent estimator. A penalized likelihood approach tends to produce a fit with fewer components, but it is not known whether that approach produces a consistent estimate of the number of mixture components. We suggest the use of a penalized minimum-distance method. It is shown that the estimator obtained is consistent for both the mixing distribution and the number of mixture components.     

A permutation test approach to the choice of size k for the nearest neighbors classifier

The k nearest neighbors (k-NN) classifier is one of the most popular methods for statistical pattern recognition and machine learning. In practice, the size k, the number of neighbors used for classification, is usually arbitrarily set to one or some other small numbers, or based on the cross-validation procedure. In this study, we propose a novel alternative approach to decide the size k. Based on a k-NN-based multivariate multi-sample test, we assign each k a permutation test based Z-score. The number of NN is set to the k with the highest Z-score. This approach is computationally efficient since we have derived the formulas for the mean and variance of the test statistic under permutation distribution for multiple sample groups. Several simulation and real-world data sets are analyzed to investigate the performance of our approach. The usefulness of our approach is demonstrated through the evaluation of prediction accuracies using Z-score as a criterion to select the size k. We also compare our approach to the widely used cross-validation approaches. The results show that the size k selected by our approach yields high prediction accuracies when informative features are used for classification, whereas the cross-validation approach may fail in some cases.     

A comparison of methods for the fitting of generalized additive models

There are several procedures for fitting generalized additive models, i.e. regression models for an exponential family response where the influence of each single covariates is assumed to have unknown, potentially non-linear shape. Simulated data are used to compare a smoothing parameter optimization approach for selection of smoothness and of covariates, a stepwise approach, a mixed model approach, and a procedure based on boosting techniques. In particular it is investigated how the performance of procedures is linked to amount of information, type of response, total number of covariates, number of influential covariates, and extent of non-linearity. Measures for comparison are prediction performance, identification of influential covariates, and smoothness of fitted functions. One result is that the mixed model approach returns sparse fits with frequently over-smoothed functions, while the functions are less smooth for the boosting approach and variable selection is less strict. The other approaches are in between with respect to these measures. The boosting procedure is seen to perform very well when little information is available and/or when a large number of covariates is to be investigated. It is somewhat surprising that in scenarios with low information the fitting of a linear model, even with stepwise variable selection, has not much advantage over the fitting of an additive model when the true underlying structure is linear. In cases with more information the prediction performance of all procedures is very similar. So, in difficult data situations the boosting approach can be recommended, in others the procedures can be chosen conditional on the aim of the analysis.     

Sequential combining in discriminant analysis

In practice, it often happens that we have a number of base methods of classification. We are not able to clearly determine which method is optimal in the sense of the smallest error rate. Then we have a combined method that allows us to consolidate information from multiple sources in a better classifier. I propose a different approach, a sequential approach. Sequentiality is understood here in the sense of adding posterior probabilities to the original data set and so created data are used during classification process. We combine posterior probabilities obtained from base classifiers using all combining methods. Finally, we combine these probabilities using a mean combining method. To the original data set we add obtained posterior probabilities as additional features. In each step we change our additional probabilities to achieve the minimum error rate for base methods. Experimental results on different data sets demonstrate that the method is efficient and that this approach outperforms base methods providing a reduction in the mean classification error rate.     

The bivariate linear model: a geometris approach

The geometric approach to the general linear model is in accessible to the majorityof statistics students be cause the computations require matrix algebra.This article presents the geometric approach for the special case of the bivariate linear model,for which the only tool require dis the in ner product.The geometric approach is introduced by showing the dual2-dimensional and5-dimensional representations of several bivariate samples x of size5.The assumptions of the bivariate model are stated geometrically,and the distributions of the regression coefficient sare derived.Theanalysis of variance(ANOVA)right triangle is pictured and the sides of the triang leare associated with their corresponding entries in the ANOVA table.     

Calculating Bivariate Orthonormal Polynomials By Recurrence

Emerson gave recurrence formulae for the calculation of orthonormal polynomials for univariate discrete random variables. He claimed that as these were based on the Christoffel–Darboux recurrence relation they were more efficient than those based on the Gram–Schmidt method. This approach was generalised by Rayner and colleagues to arbitrary univariate random variables. The only constraint was that the expectations needed are well‐defined. Here the approach is extended to arbitrary bivariate random variables for which the expectations needed are well‐defined. The extension to multivariate random variables is clear.     

A robust version of the hurdlemodel

The excess of zeros is not a rare feature in count data. Statisticians advocate the Poisson-type hurdle model (among other techniques) as an interesting approach to handle this data peculiarity. However, the frequency of gross errors and the complexity intrinsic to some considered phenomena may render this classical model unreliable and too limiting. In this paper, we develop a robust version of the Poisson hurdle model by extending the robust procedure for GLM of Cantoni and Ronchetti (2001) to the truncated Poisson regression model. The performance of the new robust approach is then investigated via a simulation study, a real data application and a sensitivity analysis. The results show the reliability of the new technique in the neighborhood of the truncated Poisson model. This robust modelling approach is therefore a valuable complement to the classical one, providing a tool for reliable statistical conclusions and to take more effective decisions.     

On the formulation of uniform laws of large numbers: a truncation approach

The paper develops a general framework for the formulation of generic uniform laws of large numbers. In particular, we introduce a basic generic uniform law of large numbers that contains recent uniform laws of large numbers by Andrews [2] and Hoadley [9] as special cases. We also develop a truncation approach that makes it possible to obtain uniform laws of large numbers for the functions under consideration from uniform laws of large numbers for truncated versions of those functions. The point of the truncation approach is that uniform laws of large numbers for the truncated versions are typically easier to obtain. By combining the basic uniform law of large numbers and the truncation approach we also derive generalizations of recent uniform laws of large numbers introduced in Pötscher and Prucha [15, 16].     

Multiple imputation of a randomly censored covariate improves logistic regression analysis

Randomly censored covariates arise frequently in epidemiologic studies. The most commonly used methods, including complete case and single imputation or substitution, suffer from inefficiency and bias. They make strong parametric assumptions or they consider limit of detection censoring only. We employ multiple imputation, in conjunction with semi-parametric modeling of the censored covariate, to overcome these shortcomings and to facilitate robust estimation. We develop a multiple imputation approach for randomly censored covariates within the framework of a logistic regression model. We use the non-parametric estimate of the covariate distribution or the semi-parametric Cox model estimate in the presence of additional covariates in the model. We evaluate this procedure in simulations, and compare its operating characteristics to those from the complete case analysis and a survival regression approach. We apply the procedures to an Alzheimer's study of the association between amyloid positivity and maternal age of onset of dementia. Multiple imputation achieves lower standard errors and higher power than the complete case approach under heavy and moderate censoring and is comparable under light censoring. The survival regression approach achieves the highest power among all procedures, but does not produce interpretable estimates of association. Multiple imputation offers a favorable alternative to complete case analysis and ad hoc substitution methods in the presence of randomly censored covariates within the framework of logistic regression.