Bayesian quadrature with non-normal approximating functions

We consider an efficient Bayesian approach to estimating integration-based posterior summaries from a separate Bayesian application. In Bayesian quadrature we model an intractable posterior density function f(·) as a Gaussian process, using an approximating function g(·), and find a posterior distribution for the integral of f(·), conditional on a few evaluations of f (·) at selected design points. Bayesian quadrature using normal g (·) is called Bayes-Hermite quadrature. We extend this theory by allowing g(·) to be chosen from two wider classes of functions. One is a family of skew densities and the other is the family of finite mixtures of normal densities. For the family of skew densities we describe an iterative updating procedure to select the most suitable approximation and apply the method to two simulated posterior density functions.     

Tables of percentage points of the k-variate normal distribution for large values of k

This paper gives tabulations of the upper α percentage points of the maximum absolute value of the k-variate normal distribution with common correlation ρ for values of k as high as 500. The tables are useful for performing multiple comparison procedures in experiments with large numbers of treatments.     

New one-stage multiple comparison procedures with the average for exponential location parameters under heteroscedasticity

ABSTRACT

In this paper, we present new one-stage multiple comparison procedures with the average for location parameters of two-parameter exponential distributions under heteroscedasticity by modifying the existing one proposed by Wu [One stage multiple comparisons with the average for exponential location parameters under heteroscedasticity. Comput Stat Data Anal. 2013;68:352–360] with unequal sample sizes. A simulation study is done and the results show that the proposed procedures have shorter confidence length with coverage probabilities closer to the nominal ones. At last, an example of comparing the survival days of patients for four categories of lung cancer is given to demonstrate the proposed procedures.     

Choosing the best pairwise comparisons of means from non-normal populations,with unequal variances,but equal sample sizes

A Monte Carlo simulation evaluated five pairwise multiple comparison procedures for controlling Type I error rates, any-pair power, and all-pairs power. Realistic conditions of non-normality were based on a previous survey. Variance ratios were varied from 1:1 to 64:1. Procedures evaluated included Tukey's honestly significant difference (HSD) preceded by an F test, the Hayter–Fisher, the Games–Howell preceded by an F test, the Pertiz with F tests, and the Peritz with Alexander–Govern tests. Tukey's procedure shows the greatest robustness in Type I error control. Any-pair power is generally best with one of the Peritz procedures. All-pairs power is best with the Pertiz F test procedure. However, Tukey's HSD preceded by the Alexander–Govern F test may provide the best combination for controlling Type I and power rates in a variety of conditions of non-normality and variance heterogeneity.     

On the statistical computation of the sample multiple correlation moefficient

Based on the recursive formulas of Lee (1988) and Singh and Relyea (1992) for computing the noncentral F distribution, a numerical algorithm for evaluating the distributional values of the sample squared multiple correlation coefficient is proposed. The distributional function of this statistic is usually represented as an infinite weighted sum of the iterative form of incomplete beta integral. So an effective algorithm for the incomplete beta integral is crucial to the numerical evaluation of various distribution values. Let a and b denote two shape parameters shown in the incomplete beta integral and hence formed in the sampling distribution functionn be the sample size, and p be the number of random variates. Then both 2a = p - 1 and 2b = n - p are positive integers in sampling situations so that the proposed numerical procedures in this paper are greatly simplified by recursively formulating the incomplete beta integral. By doing this, it can jointly compute the distributional values of probability dens function (pdf) and cumulative distribution function (cdf) for which the distributional value of quantile can be more efficiently obtained by Newton's method. In addition, computer codes in C are developed for demonstration and performance evaluation. For the less precision required, the implemented method can achieve the exact value with respect to the jnite significant digit desired. In general, the numerical results are apparently better than those by various approximations and interpolations of Gurland and Asiribo (1991),Gurland and Milton (1970), and Lee (1971, 1972). When b = (1/2)(n -p) is an integer in particular, the finite series formulation of Gurland (1968) is used to evaluate the pdf/cdf values without truncation errors, which are served as the pivotal one. By setting the implemented codes with double precisions, the infinite series form of derived method can achieve the pivotal values for almost all cases under study. Related comparisons and illustrations are also presented     

On the heterogeneity of proportions

The comparison amongmproportions can be viewed as the clustering of the means of Bernoulli trials. By introducing a distribution which is supported on the means of Bernoulli trials, we suggest a moment method approach to determine the center of the clusters. We also suggest using model selection criteria rather than the usual testing hypothesis approach to determine the grouping of the means. The discrepancy function for all possible models are compared based on the bootstrap results.     

On the effect of the number of quadrature points in a logistic random effects model: an example

Although generalized linear mixed models are recognized to be of major practical importance, it is also known that they can be computationally demanding. The problem is the evaluation of the integral in calculating the marginalized likelihood. The straightforward method is based on the Gauss–Hermite technique, based on Gaussian quadrature points. Another approach is provided by the class of penalized quasi-likelihood methods. It is commonly believed that the Gauss–Hermite method works relatively well in simple situations but fails in more complicated structures. However, we present here a strikingly simple example of a logistic random-intercepts model in the context of a longitudinal clinical trial where the method gives valid results only for a high number of quadrature points ( Q ). As a consequence, this result warns the practitioner to examine routinely the dependence of the results on Q . The adaptive Gaussian quadrature, as implemented in the new SAS procedure NLMIXED, offered the solution to our problem. However, even the adaptive version of Gaussian quadrature needs careful handling to ensure convergence.     

Multiple comparison for checking differences among normal variances

In this study, we discuss multiple comparisons for checking differences among normal variances. Specifically, we consider the multiple comparison with a control and the all-pairwise multiple comparison. For the multiple comparison with a control we determine the critical value for pairwise comparison satisfying a specified significance level exactly and formulate the power of the test under a specified alternative hypothesis. For the all-pairwise multiple comparison we determine two kinds of conservative critical values for pairwise comparison for a specified significance level and calculate the power of the test by Monte Carlo simulation. Finally, we give some numerical examples regarding critical values and power of the test.     

Testing non-inferiority and superiority for two endpoints for several treatments with a control

Some multiple comparison procedures are described for multiple armed studies. The procedures are appropriate for testing all hypotheses for comparing two endpoints and multiple test arms to a single control group, for example three different fixed doses compared to a placebo. The procedure assumes that among the two endpoints, one is designated as a primary endpoint such that for a given treatment arm, no hypothesis for the secondary endpoint can be rejected unless the hypothesis for the primary endpoint was rejected. The procedures described control the family-wise error rate in the strong sense at a specified level α.     

Robust inference from multiple test statistics via permutations: a better alternative to the single test statistic approach for randomized trials

Formal inference in randomized clinical trials is based on controlling the type I error rate associated with a single pre‐specified statistic. The deficiency of using just one method of analysis is that it depends on assumptions that may not be met. For robust inference, we propose pre‐specifying multiple test statistics and relying on the minimum p‐value for testing the null hypothesis of no treatment effect. The null hypothesis associated with the various test statistics is that the treatment groups are indistinguishable. The critical value for hypothesis testing comes from permutation distributions. Rejection of the null hypothesis when the smallest p‐value is less than the critical value controls the type I error rate at its designated value. Even if one of the candidate test statistics has low power, the adverse effect on the power of the minimum p‐value statistic is not much. Its use is illustrated with examples. We conclude that it is better to rely on the minimum p‐value rather than a single statistic particularly when that single statistic is the logrank test, because of the cost and complexity of many survival trials. Copyright © 2013 John Wiley & Sons, Ltd.     

TWO-WAY CHANGE-POINT MODEL AND ITS APPLICATION

Exact null and alternative distributions of the two-way maximally selected x² for interaction between the ordered rows and columns are derived for each of the normal and Poisson models, respectively. The method is one of the multiple comparison procedures for ordered parameters and is useful for defining a block interaction or a two-way change-point model as a simple alternative to the two-way additive model. The construction of a confidence region for the two-way change-point is then described. An important application is found in a dose-response clinical trial with ordered categorical responses, where detecting the dose level which gives significantly higher responses than the lower doses can be formulated as a problem of detecting a change in the interaction effects.     

A method of unconstrained multiple comparisons with the best

There are three types of multiple comparisons: all-pairwise multiple comparisons (MCA), multiple comparisons with the best (MCB), and multiple comparisons with a control (MCC). There are also three levels of multiple comparisons inference: confidence sets, subset comparisons, test of homogeneity. In current practice, MCA procedures dominate. In correct attempts at more efficient comparisons, in the form of employing lower level MCA procedures for higher level inference, account for the most frequent abuses in multiple comparisons. A better strategy is to choose the correct type of inference at the level of inference desired. In particular, very often the simulataneous comparisons of each treatment with the best of the other treatments (MCB) suffice. Hsu (1984b) gave simultaneous confidence intervals for θ_i ? max_j≠iθ_j having the simple form [? (Y_i ?max_j≠i Y_j ? C) (Y_i?max_j≠i Y_j + C)⁺]. Those intervals were constrained, sothat even if a treatment is inferred to be the best, no positive bound on how much it is better thatn the rest is given, a somewhat undesirable property. In this article it is shown that by employing a slightly larger critical value, the nonpositivity constraint on the lower bound is removed.     

Change PP plot and continous sample quantile function

Functional inference recommends data analysis of a sample of n observations by functional and graphical representations of its probability models using various functions on 0 < u < 1, including the quantile function. This paper discusses: charge PP plots and a continuous version of the sample quantile function which use the mid-distinct value probability integral transform; comparison density functions; comparison interpretation of probability integral transform; maximum spacings method of one sample parameter estimation.     

On the Critical Value Behaviour of Multiple Decision Procedures

In this paper we investigate the asymptotic critical value behaviour of certain multiple decision procedures as e.g. simultaneous confidence intervals and simultaneous as well as stepwise multiple test procedures. Supposing that n hypotheses or parameters of interest are under consideration we investigate the critical value behaviour when n increases. More specifically, we answer e.g. the question by which amount the lengths of confidence intervals increase when an additional parameter is added to the statistical analysis. Furthermore, critical values of different multiple decision procedures as for instance step-down and step-up procedures will be compared. Some general theoretic results are derived and applied for various distributions.     

Multiple comparisons of several populations with more than one control with respectto scale parameter

In this paper, one-sided and two-sided test procedures for comparing several treatments with more than one control with respect to scale parameter are proposed. The proposed test procedures are inverted to obtain the associated simultaneous confidence intervals. The multiple comparisons of test treatments with the best control are also developed. The computation of the critical points, required to implement the proposed procedures, is discussed by taking the normal probability model. Applications of the proposed test procedures to two-parameter exponential probability model are also demonstrated.     

Exponential Inequalities in Stochastic Inverse Problems Using an Iterative Method

We consider an iterative method in order to solve linear inverse problems. We establish exponential inequalities for the probability of the distance between the approximated solution and the exact one for a calibration problem. The approximate is given by an iterative method with Gaussian errors. We treat an operator equation of the form Ax = u, where A is a compact operator.     

Nonparametric approaches to testing and multiple comparisons in a one-way anova

Two questions of interest involving nonparametric multiple comparisons are considered. The first question concerns whether it is appropriate to use a multiple comparison procedure as a test of the equality of k treatments, and if it is, which procedure performs best as a test. Our results show that for smaller k values some multiple comparison procedures perform well as tests. The second question concerns whether a joint ranking or a separate ranking multiple comparison procedure performs better as a test and as a device for treatment separation. We find that the joint ranking procedure does slightly better as a test, but for treatment separation the answer depends on the situation.     

Performance of likelihood-based estimation methods for multilevel binary regression models

By means of a fractional factorial simulation experiment, we compare the performance of penalised quasi-likelihood (PQL), non-adaptive Gaussian quadrature and adaptive Gaussian quadrature in estimating parameters for multilevel logistic regression models. The comparison is done in terms of bias, mean-squared error (MSE), numerical convergence and computational efficiency. It turns out that in terms of MSE, standard versions of the quadrature methods perform relatively poorly in comparison with PQL.     

Accurate and efficient numerical calculation of stable densities via optimized quadrature and asymptotics

Stable distributions are an important class of infinitely divisible probability distributions, of which two special cases are the Cauchy distribution and the normal distribution. Aside from a few special cases, the density function for stable distributions has no known analytic form and is expressible only through the variate’s characteristic function or other integral forms. In this paper, we present numerical schemes for evaluating the density function for stable distributions, its gradient, and distribution function in various parameter regimes of interest, some of which had no preexisting efficient method for their computation. The novel evaluation schemes consist of optimized generalized Gaussian quadrature rules for integral representations of the density function, complemented by asymptotic expansions near various values of the shape and argument parameters. We report several numerical examples illustrating the efficiency of our methods. The resulting code has been made available online.     

Multiple comparisons with the average for exponential location parameters based on doubly censored sample under heteroscedasticity

In this paper, a one-stage multiple comparison procedures with the average for exponential location parameters based on the doubly censored sample under heteroscedasticity is proposed. These intervals can be used to identify a subset which includes all no-worse-than-the-average treatments in an experimental design and to identify better-than-the-average, worse-than-the-average and not-much-different-from-the-average products in agriculture, emerging market, pharmaceutical industries. The critical values are tabulated in a table for practical use. A simulation study on the confidence length and coverage probabilities is done. At last, an example of comparing four drugs in the treatment of leukemia is given to demonstrate the proposed procedures.