Saddlepoint approximations for some models of circular data

In this article we provide saddlepoint approximations for some important models of circular data. The particularity of these saddlepoint approximations is that they do not require solving the saddlepoint equation iteratively, so their evaluation is immediate. We first give very accurate approximations to P-values, critical values and power functions for some optimal tests regarding the concentration parameter under wrapped symmetric α-stable and circular normal models. Then, we consider an approximation to the distribution of a projection of the two-dimensional Pearson random walk with exponential step sizes.     

An Examination of Discrepancies in Multiple Imputation Procedures Between SAS® and SPSS®

Multiple imputation (MI) has become a feasible method to replace missing data due to the rapid development of computer technology over the past three decades. Nonetheless, a unique issue with MI hinges on the fact that different software packages can give different results. Even when one begins with the same random number seed, conflicting findings can be obtained from the same data under an identical imputation model between SAS® and SPSS®. Consequently, as illustrated in this article, a predictor variable can be claimed both significant and not significant depending on the software being used. Based on the considerations of multiple imputation steps, including result pooling, default selection, and different numbers of imputations, practical suggestions are provided to minimize the discrepancies in the results obtained when using MI. Features of Stata® are briefly reviewed in the Discussion section to broaden the comparison of MI computing across widely used software packages.     

On the computation of the noncentral <Emphasis Type="Italic">F</Emphasis> and noncentral beta distribution

Unfortunately many of the numerous algorithms for computing the comulative distribution function (cdf) and noncentrality parameter of the noncentral F and beta distributions can produce completely incorrect results as demonstrated in the paper by examples. Existing algorithms are scrutinized and those parts that involve numerical difficulties are identified. As a result, a pseudo code is presented in which all the known numerical problems are resolved. This pseudo code can be easily implemented in programming language C or FORTRAN without understanding the complicated mathematical background. Symbolic evaluation of a finite and closed formula is proposed to compute exact cdf values. This approach makes it possible to check quickly and reliably the values returned by professional statistical packages over an extraordinarily wide parameter range without any programming knowledge. This research was motivated by the fact that a very useful table for calculating the size of detectable effects for ANOVA tables contains suspect values in the region of large noncentrality parameter values compared to the values obtained by Patnaik’s 2-moment central-F approximation. The cause is identified and the corrected form of the table for ANOVA purposes is given. The accuracy of the approximations to the noncentral-F distribution is also discussed. The authors wish to thank Mr. Richárd Király for his preliminary work. The authors are grateful to the Editor and Associate Editor of STCO and the unknown reviewers for their helpful suggestions.     

Computing factors for exact two-sided tolerance limits for a normal distribution

A self-contained FORTRAN subroutine is provided which computes factors for Wald-Wolfowitz type tolerance limits allowing arbitrary combinations of sample size n and degrees of freedom ν. The exact calculations from our program reveal inadequacies of two existing approximations, especially when ν ? n. Numerous applications where ν ≠ n ? 1 are cited; two of these are discussed and illustrated.     

Heteroscedastic ANOVA: old p values, new views

The generalization of the Behrens–Fisher problem to comparing more than two means from nonhomogeneous populations has attracted the attention of statisticians for many decades. Several approaches offer different approximations to the distribution of the test statistic. The question of statistical properties of these approximations is still alive. Here, we present a brief overview of several approaches suggested in the literature and implemented in software with a focus on investigating the accuracy of p values as well as their dependence on nuisance parameters and on the underlying assumption of normality. We illustrate by simulation the behavior of p values. In addition to the Satterthwaite–Fai–Cornelius test, the Kenward–Roger test, the simple ANOVA F test, the parametric bootstrap test, and the generalized F test will be briefly discussed.     

Log‐rank permutation tests for trend: saddlepoint p‐values and survival rate confidence intervals

Suppose p + 1 experimental groups correspond to increasing dose levels of a treatment and all groups are subject to right censoring. In such instances, permutation tests for trend can be performed based on statistics derived from the weighted log‐rank class. This article uses saddlepoint methods to determine the mid‐P‐values for such permutation tests for any test statistic in the weighted log‐rank class. Permutation simulations are replaced by analytical saddlepoint computations which provide extremely accurate mid‐P‐values that are exact for most practical purposes and almost always more accurate than normal approximations. The speed of mid‐P‐value computation allows for the inversion of such tests to determine confidence intervals for the percentage increase in mean (or median) survival time per unit increase in dosage. The Canadian Journal of Statistics 37: 5‐16; 2009 © 2009 Statistical Society of Canada     

Computing Critical Values of Exact Tests by Incorporating Monte Carlo Simulations Combined with Statistical Tables

Various exact tests for statistical inference are available for powerful and accurate decision rules provided that corresponding critical values are tabulated or evaluated via Monte Carlo methods. This article introduces a novel hybrid method for computing p‐values of exact tests by combining Monte Carlo simulations and statistical tables generated a priori. To use the data from Monte Carlo generations and tabulated critical values jointly, we employ kernel density estimation within Bayesian‐type procedures. The p‐values are linked to the posterior means of quantiles. In this framework, we present relevant information from the Monte Carlo experiments via likelihood‐type functions, whereas tabulated critical values are used to reflect prior distributions. The local maximum likelihood technique is employed to compute functional forms of prior distributions from statistical tables. Empirical likelihood functions are proposed to replace parametric likelihood functions within the structure of the posterior mean calculations to provide a Bayesian‐type procedure with a distribution‐free set of assumptions. We derive the asymptotic properties of the proposed nonparametric posterior means of quantiles process. Using the theoretical propositions, we calculate the minimum number of needed Monte Carlo resamples for desired level of accuracy on the basis of distances between actual data characteristics (e.g. sample sizes) and characteristics of data used to present corresponding critical values in a table. The proposed approach makes practical applications of exact tests simple and rapid. Implementations of the proposed technique are easily carried out via the recently developed STATA and R statistical packages.     

Power and Sample Size for Fixed-Effects Inference in Reversible Linear Mixed Models

ABSTRACT

Despite the popularity of the general linear mixed model for data analysis, power and sample size methods and software are not generally available for commonly used test statistics and reference distributions. Statisticians resort to simulations with homegrown and uncertified programs or rough approximations which are misaligned with the data analysis. For a wide range of designs with longitudinal and clustering features, we provide accurate power and sample size approximations for inference about fixed effects in the linear models we call reversible. We show that under widely applicable conditions, the general linear mixed-model Wald test has noncentral distributions equivalent to well-studied multivariate tests. In turn, exact and approximate power and sample size results for the multivariate Hotelling–Lawley test provide exact and approximate power and sample size results for the mixed-model Wald test. The calculations are easily computed with a free, open-source product that requires only a web browser to use. Commercial software can be used for a smaller range of reversible models. Simple approximations allow accounting for modest amounts of missing data. A real-world example illustrates the methods. Sample size results are presented for a multicenter study on pregnancy. The proposed study, an extension of a funded project, has clustering within clinic. Exchangeability among the participants allows averaging across them to remove the clustering structure. The resulting simplified design is a single-level longitudinal study. Multivariate methods for power provide an approximate sample size. All proofs and inputs for the example are in the supplementary materials (available online).     

Guidelines for accurate EC50/IC50 estimation

This article provides minimum requirements for having confidence in the accuracy of EC50/IC50 estimates. Two definitions of EC50/IC50s are considered: relative and absolute. The relative EC50/IC50 is the parameter c in the 4-parameter logistic model and is the concentration corresponding to a response midway between the estimates of the lower and upper plateaus. The absolute EC50/IC50 is the response corresponding to the 50% control (the mean of the 0% and 100% assay controls). The guidelines first describe how to decide whether to use the relative EC50/IC50 or the absolute EC50/IC50. Assays for which there is no stable 100% control must use the relative EC50/IC50. Assays having a stable 100% control but for which there may be more than 5% error in the estimate of the 50% control mean should use the relative EC50/IC50. Assays that can be demonstrated to produce an accurate and stable 100% control and less than 5% error in the estimate of the 50% control mean may gain efficiency as well as accuracy by using the absolute EC50/IC50. Next, the guidelines provide rules for deciding when the EC50/IC50 estimates are reportable. The relative EC50/IC50 should only be used if there are at least two assay concentrations beyond the lower and upper bend points. The absolute EC50/IC50 should only be used if there are at least two assay concentrations whose predicted response is less than 50% and two whose predicted response is greater than 50%. A wide range of typical assay conditions are considered in the development of the guidelines.     

The exact distribution of the maximum,minimum and the range of Multinomial/Dirichlet and Multivariate Hypergeometric frequencies

The exact distribution of the maximum and minimum frequencies of Multinomial/Dirichlet and Multivariate Hypergeometric distributions of n balls in m urns is compactly represented as a product of stochastic matrices. This representation does not require equal urn probabilities, is invariant to urn order, and permits rapid calculation of exact probabilities. The exact distribution of the range is also obtained. These algorithms satisfy a long-standing need for routines to compute exact Multinomial/Dirichlet and Multivariate Hypergeometric maximum, minimum, and range probabilities in statistical computation libraries and software packages.     

The Mean,Median, and Confidence Intervals of the Kaplan-Meier Survival Estimate—Computations and Applications

This short note points out estimators of the mean, median, and the associated confidence intervals of the Kaplan-Meier product limit estimate. Some uses of the estimator of the mean are described. In addition, differences among popular software packages in the calculation of both the mean and median and associated confidence intervals are demonstrated and are due to default settings in the software. Simple examples of the calculations are provided using S-Plus, R, SAS, Stata, and SPSS.     

Sunset Salvo

The Wilcoxon—Mann—Whitney test enjoys great popularity among scientists comparing two groups of observations, especially when measurements made on a continuous scale are non-normally distributed. Triggered by different results for the procedure from two statistics programs, we compared the outcomes from 11 PC-based statistics packages. The findings were that the delivered p values ranged from significant to nonsignificant at the 5% level, depending on whether a large-sample approximation or an exact permutation form of the test was used and, in the former case, whether or not a correction for continuity was used and whether or not a correction for ties was made. Some packages also produced pseudo-exact p values, based on the null distribution under the assumption of no ties. A further crucial point is that the variant of the algorithm used for computation by the packages is rarely indicated in the output or documented in the Help facility and the manuals. We conclude that the only accurate form of the Wilcoxon—Mann—Whitney procedure is one in which the exact permutation null distribution is compiled for the actual data.     

Saddlepoint p-values for a class of tests for comparing competing risks with censored data

One of the general problems in clinical trials and mortality rates is the comparison of competing risks. Most of the test statistics used for independent and dependent risks with censored data belong to the class of weighted linear rank tests in its multivariate version. In this paper, we introduce the saddlepoint approximations as accurate and fast approximations for the exact p-values of this class of tests instead of the asymptotic and permutation simulated calculations. Real data examples and extensive simulation studies showed the accuracy and stability performance of the saddlepoint approximations over different scenarios of lifetime distributions, sample sizes and censoring.     

Comparison of sample size formulae for 2 × 2 cross‐over designs applied to bioequivalence studies

We consider the comparison of two formulations in terms of average bioequivalence using the 2 × 2 cross‐over design. In a bioequivalence study, the primary outcome is a pharmacokinetic measure, such as the area under the plasma concentration by time curve, which is usually assumed to have a lognormal distribution. The criterion typically used for claiming bioequivalence is that the 90% confidence interval for the ratio of the means should lie within the interval (0.80, 1.25), or equivalently the 90% confidence interval for the differences in the means on the natural log scale should be within the interval (?0.2231, 0.2231). We compare the gold standard method for calculation of the sample size based on the non‐central t distribution with those based on the central t and normal distributions. In practice, the differences between the various approaches are likely to be small. Further approximations to the power function are sometimes used to simplify the calculations. These approximations should be used with caution, because the sample size required for a desirable level of power might be under‐ or overestimated compared to the gold standard method. However, in some situations the approximate methods produce very similar sample sizes to the gold standard method. Copyright © 2005 John Wiley & Sons, Ltd.     

Covariance Analysis for Split-Plot and Split-Block Designs

Commentaries are informative essays dealing with viewpoints of statistical practice, statistical education, and other topics considered to be of general interest to the broad readership of The American Statistician. Commentaries are similar in spirit to Letters to the Editor, but they involve longer discussions of background, issues, and perspectives. All commentaries will be refereed for their merit and compatibility with these criteria.

Proper methodology for the analysis of covariance for experiments designed in a split-plot or split-block design is not found in the statistical literature. Analyses for these designs are often performed incompletely or even incorrectly. This is especially true when popular statistical computer software packages are used for the analysis of these designs. This article provides several appropriate models, ANOVA tables, and standard errors for comparisons from experiments arranged in a standard split-plot, split–split-plot, or split-block design where a covariate has been measured on the smallest size experimental unit.     

Asymptotic behaviour of M‐estimators in AR(p) models under nonstandard conditions

The authors derive the limiting distribution of M‐estimators in AR(p) models under nonstandard conditions, allowing for discontinuities in score and density functions. Unlike usual regularity assumptions, these conditions are satisfied in the context of L₁‐estimation and autoregression quantiles. The asymptotic distributions of the resulting estimators, however, are not generally Gaussian. Moreover, their bootstrap approximations are consistent along very specific sequences of bootstrap sample sizes only.     

Notes on Kolassa’s Method for Estimating the Power of Wilcoxon–Mann–Whitney Test

The Kolassa method implemented in the nQuery Advisor software has been widely used for approximating the power of the Wilcoxon–Mann–Whitney (WMW) test for ordered categorical data, in which Edgeworth approximation is used to estimate the power of an unconditional test based on the WMW U statistic. When the sample size is small or when the sizes in the two groups are unequal, Kolassa’s method may yield quite poor approximation to the power of the conditional WMW test that is commonly implemented in statistical packages. Two modifications of Kolassa’s formula are proposed and assessed by simulation studies.     

Calculating the density and distribution function for the singly and doubly noncentral F

Simple, closed form saddlepoint approximations for the distribution and density of the singly and doubly noncentral F distributions are presented. Their overwhelming accuracy is demonstrated numerically using a variety of parameter values. The approximations are shown to be uniform in the right tail and the associated limitating relative error is derived. Difficulties associated with some algorithms used for exact computation of the singly noncentral F are noted.     

Meta-Analysis Software: A Comparative Review

DSTAT, Version 1.10: Available from Lawrence Erlbaum Associates, Inc., 10 Industrial Ave., Mahwah, NJ 07430-2262; phone: 800-926-6579

TRUE EPISTAT, Version 4.0: Available from Epistat Services, 2011 Cap Rock Circle, Richardson, TX 75080; phone: 214-680-1376; fax: 214-680-1303.

FAST*PRO, Version 1.0: Available from Academic Press, Inc., 955 Massachusetts Avenue, Cambridge, MA 02139; phone: 800-321-5068; fax: 800-336-7377.

Meta-analysts conduct studies in which the responses are analytic summary measurements, such as risk differences, effect sizes, p values, or z statistics, obtained from a series of independent studies. The motivation for conducting a meta-analysis is to integrate research findings over studies in order to summarize the evidence about treatment efficacy or risk factors. This article presents a comparative review of three meta-analytic software packages: DSTAT, TRUE EPISTAT, and FAST*PRO.     

Testing linear hypotheses in high-dimensional regressions

For a multivariate linear model, Wilk's likelihood ratio test (LRT) constitutes one of the cornerstone tools. However, the computation of its quantiles under the null or the alternative hypothesis requires complex analytic approximations, and more importantly, these distributional approximations are feasible only for moderate dimension of the dependent variable, say p≤20. On the other hand, assuming that the data dimension p as well as the number q of regression variables are fixed while the sample size n grows, several asymptotic approximations are proposed in the literature for Wilk's Λ including the widely used chi-square approximation. In this paper, we consider necessary modifications to Wilk's test in a high-dimensional context, specifically assuming a high data dimension p and a large sample size n. Based on recent random matrix theory, the correction we propose to Wilk's test is asymptotically Gaussian under the null hypothesis and simulations demonstrate that the corrected LRT has very satisfactory size and power, surely in the large p and large n context, but also for moderately large data dimensions such as p=30 or p=50. As a byproduct, we give a reason explaining why the standard chi-square approximation fails for high-dimensional data. We also introduce a new procedure for the classical multiple sample significance test in multivariate analysis of variance which is valid for high-dimensional data.