Fisher's Hypergeometric Test for a Comparison in a Finite Population

Graphical methods have played a central role in the development of statistical theory and practice. This presentation briefly reviews some of the highlights in the historical development of statistical graphics and gives a simple taxonomy that can be used to characterize the current use of graphical methods. This taxonomy is used to describe the evolution of the use of graphics in some major statistical and related scientific journals.

Some recent advances in the use of graphical methods for statistical analysis are reviewed, and several graphical methods for the statistical presentation of data are illustrated, including the use of multicolor maps.     

An Alternative to Transfer Function Forecasting Based on Subspace Methods

In the time series literature, recent interest has focused on the so-called subspace methods. These techniques use canonical correlations and linear regressions to estimate the system matrices of an ARMAX model expressed in state space form. In this article, we use subspace methods to forecast two series with the help of some exogenous variables related to them. We compare the results with those obtained using traditional transfer function models and find that the forecasts obtained with both methods are similar. This result is very encouraging because, in contrast to transfer function models, subspace methods can be considered as almost automatic.     

Mixture Kalman filters

In treating dynamic systems, sequential Monte Carlo methods use discrete samples to represent a complicated probability distribution and use rejection sampling, importance sampling and weighted resampling to complete the on-line 'filtering' task. We propose a special sequential Monte Carlo method, the mixture Kalman filter, which uses a random mixture of the Gaussian distributions to approximate a target distribution. It is designed for on-line estimation and prediction of conditional and partial conditional dynamic linear models, which are themselves a class of widely used non-linear systems and also serve to approximate many others. Compared with a few available filtering methods including Monte Carlo methods, the gain in efficiency that is provided by the mixture Kalman filter can be very substantial. Another contribution of the paper is the formulation of many non-linear systems into conditional or partial conditional linear form, to which the mixture Kalman filter can be applied. Examples in target tracking and digital communications are given to demonstrate the procedures proposed.     

New High-Speed Computing Devices

“One method of error analysis (not the one we will use) is based upon the principles of mathematical statistics. Unfortunately, statistical methods can only be meaningfully applied when one has large amounts of data for a given system. In many cases … these large quantities of data are not available … then statistical methods are not applicable, and some other methods must be devised.”     

Automatic breakpoint detection for retrospective cumulative sum charts

Cumulative sum (cusum) methods can be used for monitoring processes and for retrospective (historical) data analysis. Most software only provides the former. The comment by Williamson that retrospective cusum analysis is a neglected area is still true. Though not in vogue, retrospective cusum analysis is useful for investigations such as benchmarking of processes, identifying causes of process decay, selecting reference data sets for typicality studies, and reporting of historical data. Even those texts which cover retrospective analyses, usually ignore the question of identifying multiple points of change (breakpoints), and present essentially manual methods for assessing single breakpoints. Most users of statistical methods want software solutions that are easy to use and require little user intervention or interpretation. Direct implementation of manual method does not give a user robust solution. Problems are illustrated. Attempts to use monitoring CuSums in retrospective analysis can also lead to errors. A practical recursive method is presented for breakpoint identification and significance assessment, which can be automated. Copyright © 2002 John Wiley & Sons, Ltd.     

Likelihood Maximization Using Web-Based Optimization Tools

This article proposes the use of optimization techniques and tools to maximize the likelihood if maximization cannot be easily accomplished with standard statistical software. In such situations, the use of the programming language AMPL with the freely available optimization solvers under the NEOS Server is an attractive alternative to algorithms developed for specific optimization problems in statistics. This article is meant to be a short tutorial introducing statisticians to these methods and tools. We provide an example to illustrate these methods. The necessary files for maximization are included in the Appendix so that the reader can carry out the optimization procedure described.     

Finding 'superclassifications' with an acceptable misclassification rate

Cluster analysis methods are based on measures of 'distance' between objects. Sometimes the objects have an internal structure, and use of this can be made when defining such distances. This leads to non-standard cluster analysis methods. We illustrate with an application in which the objects are themselves classes and the aim is to produce clusters of classes which minimize the error rate of a supervised classification rule. For supervised classification problems with more than a handful of classes, there may exist groups of classes which are well separated from other groups, even though individual classes are not all well separated. In such cases, the overall misclassification rate is a crude measure of performance and more subtle measures, taking note of subgroup separation, are desirable. The fact that points can be assigned accurately to groups, if not to individual classes, can sometimes be practically useful.     

The use of simple reparameterizations to improve the efficiency of Markov chain Monte Carlo estimation for multilevel models with applications to discrete time survival models

Summary.  We consider the application of Markov chain Monte Carlo (MCMC) estimation methods to random-effects models and in particular the family of discrete time survival models. Survival models can be used in many situations in the medical and social sciences and we illustrate their use through two examples that differ in terms of both substantive area and data structure. A multilevel discrete time survival analysis involves expanding the data set so that the model can be cast as a standard multilevel binary response model. For such models it has been shown that MCMC methods have advantages in terms of reducing estimate bias. However, the data expansion results in very large data sets for which MCMC estimation is often slow and can produce chains that exhibit poor mixing. Any way of improving the mixing will result in both speeding up the methods and more confidence in the estimates that are produced. The MCMC methodological literature is full of alternative algorithms designed to improve mixing of chains and we describe three reparameterization techniques that are easy to implement in available software. We consider two examples of multilevel survival analysis: incidence of mastitis in dairy cattle and contraceptive use dynamics in Indonesia. For each application we show where the reparameterization techniques can be used and assess their performance.     

Parametric Estimation Procedures in Multivariate Generalized Pareto Models

Abstract.  Modelling the tails of a multivariate distribution can be reasonably done by multivariate generalized Pareto distributions (GPDs). We present several methods of parametric estimation in these models, which use decompositions of the corresponding random vectors with the help of different versions of Pickands coordinates. The estimators are compared to each other with simulated data sets. To show the practical value of the methods, they are applied to a real hydrological data set.     

ML, PL, QL in Markov Chain Models

Abstract.  In many spatial and spatial-temporal models, and more generally in models with complex dependencies, it may be too difficult to carry out full maximum-likelihood (ML) analysis. Remedies include the use of pseudo-likelihood (PL) and quasi-likelihood (QL) (also called the composite likelihood). The present paper studies the ML, PL and QL methods for general Markov chain models, partly motivated by the desire to understand the precise behaviour of the PL and QL methods in settings where this can be analysed. We present limiting normality results and compare performances in different settings. For Markov chain models, the PL and QL methods can be seen as maximum penalized likelihood methods. We find that QL is typically preferable to PL, and that it loses very little to ML, while sometimes earning in model robustness. It has also appeal and potential as a modelling tool. Our methods are illustrated for consonant-vowel transitions in poetry and for analysis of DNA sequence evolution-type models.     

Mixture Distributions Based Methods of Calibration for the Empirical Log-Likelihood Ratio

Empirical likelihood ratio confidence regions based on the chi-square calibration suffer from an undercoverage problem in that their actual coverage levels tend to be lower than the nominal levels. The finite sample distribution of the empirical log-likelihood ratio is recognized to have a mixture structure with a continuous component on [0, + ∞) and a point mass at + ∞. The undercoverage problem of the Chi-square calibration is partly due to its use of the continuous Chi-square distribution to approximate the mixture distribution of the empirical log-likelihood ratio. In this article, we propose two new methods of calibration which will take advantage of the mixture structure; we construct two new mixture distributions by using the F and chi-square distributions and use these to approximate the mixture distributions of the empirical log-likelihood ratio. The new methods of calibration are asymptotically equivalent to the chi-square calibration. But the new methods, in particular the F mixture based method, can be substantially more accurate than the chi-square calibration for small and moderately large sample sizes. The new methods are also as easy to use as the chi-square calibration.     

Function-on-function regression for two-dimensional functional data

ABSTRACT

We present methods for modeling and estimation of a concurrent functional regression when the predictors and responses are two-dimensional functional datasets. The implementations use spline basis functions and model fitting is based on smoothing penalties and mixed model estimation. The proposed methods are implemented in available statistical software, allow the construction of confidence intervals for the bivariate model parameters, and can be applied to completely or sparsely sampled responses. Methods are tested to data in simulations and they show favorable results in practice. The usefulness of the methods is illustrated in an application to environmental data.     

Robust methods for personal-income distribution models

Statistical problems in modelling personal-income distributions include estimation procedures, testing, and model choice. Typically, the parameters of a given model are estimated by classical procedures such as maximum-likelihood and least-squares estimators. Unfortunately, the classical methods are very sensitive to model deviations such as gross errors in the data, grouping effects, or model misspecifications. These deviations can ruin the values of the estimators and inequality measures and can produce false information about the distribution of the personal income in a country. In this paper we discuss the use of robust techniques for the estimation of income distributions. These methods behave like the classical procedures at the model but are less influenced by model deviations and can be applied to general estimation problems.     

Measuring the Precision of Statistical Cost Allocations

This article examines the use of regression analysis for allocating indirect costs. When multiple regression is used to estimate the weights of several allocation factors, conventional standard errors and correlation coefficients can be misleading with respect to the statistical precision of the cost allocations. This article develops alternative allocation approaches and measures of precision that use linear prediction theory and Bayesian inference. The proposed methods are illustrated using a university indirect cost study.     

Bayesian model comparison with un-normalised likelihoods

Models for which the likelihood function can be evaluated only up to a parameter-dependent unknown normalizing constant, such as Markov random field models, are used widely in computer science, statistical physics, spatial statistics, and network analysis. However, Bayesian analysis of these models using standard Monte Carlo methods is not possible due to the intractability of their likelihood functions. Several methods that permit exact, or close to exact, simulation from the posterior distribution have recently been developed. However, estimating the evidence and Bayes’ factors for these models remains challenging in general. This paper describes new random weight importance sampling and sequential Monte Carlo methods for estimating BFs that use simulation to circumvent the evaluation of the intractable likelihood, and compares them to existing methods. In some cases we observe an advantage in the use of biased weight estimates. An initial investigation into the theoretical and empirical properties of this class of methods is presented. Some support for the use of biased estimates is presented, but we advocate caution in the use of such estimates.     

Covariate Decomposition Methods for Longitudinal Missing‐at‐Random Data and Predictors Associated with Subject‐Specific Effects

Investigators often gather longitudinal data to assess changes in responses over time within subjects and to relate these changes to within‐subject changes in predictors. Missing data are common in such studies and predictors can be correlated with subject‐specific effects. Maximum likelihood methods for generalized linear mixed models provide consistent estimates when the data are ‘missing at random’ (MAR) but can produce inconsistent estimates in settings where the random effects are correlated with one of the predictors. On the other hand, conditional maximum likelihood methods (and closely related maximum likelihood methods that partition covariates into between‐ and within‐cluster components) provide consistent estimation when random effects are correlated with predictors but can produce inconsistent covariate effect estimates when data are MAR. Using theory, simulation studies, and fits to example data this paper shows that decomposition methods using complete covariate information produce consistent estimates. In some practical cases these methods, that ostensibly require complete covariate information, actually only involve the observed covariates. These results offer an easy‐to‐use approach to simultaneously protect against bias from both cluster‐level confounding and MAR missingness in assessments of change.     

Why should clinicians care about Bayesian methods?

There is a growing awareness of Bayesian methods within the medical research community, and increasing discussion of their potential applications. This interest has, however, so far failed to convert into the routine use of such methods by working clinicians. I argue that attempts to encourage the use of Bayesian methods by highlighting the deficiencies of conventional (frequentist) inference have not succeeded because these deficiencies typically have minor practical consequences, while their more serious effects can usually be explained away by appeal to other statistical issues. As a result, Bayesian methods have not appeared to offer practical pay-offs big enough to justify the cost of acquiring the necessary expertise. In an attempt to remove this “cost–benefit” hurdle, I outline a simple Bayesian technique that can be used alongside frequentist methods to address an issue of routine practical concern to working clinicians: the credibility of new research findings.     

Estimation of Non‐Crossing Quantile Regression Curves

Quantile regression methods have been widely used in many research areas in recent years. However conventional estimation methods for quantile regression models do not guarantee that the estimated quantile curves will be non‐crossing. While there are various methods in the literature to deal with this problem, many of these methods force the model parameters to lie within a subset of the parameter space in order for the required monotonicity to be satisfied. Note that different methods may use different subspaces of the space of model parameters. This paper establishes a relationship between the monotonicity of the estimated conditional quantiles and the comonotonicity of the model parameters. We develope a novel quasi‐Bayesian method for parameter estimation which can be used to deal with both time series and independent statistical data. Simulation studies and an application to real financial returns show that the proposed method has the potential to be very useful in practice.     

Bayesian estimation methods for categorical data with misclassifications

This article considers Bayesian estimation methods for categorical data with misclassifications. To adjust for misclassification, double sampling schemes are utilized. Observations are represented in a contingency table categorized by error-free categorical variables and error-prone categorical variables. Posterior means of probabilities in cells are considered as estimates. In some cases, the posterior means can be calculated exactly. However,in some cases, the exact calculation may be too difficult to perform, but we can easily use the expectation-maximiza-tion(EM) algorithm to obtain approximate posterior means.     

Non-parametric Estimation for NHPP Software Reliability Models

The non-homogeneous Poisson process (NHPP) model is a very important class of software reliability models and is widely used in software reliability engineering. NHPPs are characterized by their intensity functions. In the literature it is usually assumed that the functional forms of the intensity functions are known and only some parameters in intensity functions are unknown. The parametric statistical methods can then be applied to estimate or to test the unknown reliability models. However, in realistic situations it is often the case that the functional form of the failure intensity is not very well known or is completely unknown. In this case we have to use functional (non-parametric) estimation methods. The non-parametric techniques do not require any preliminary assumption on the software models and then can reduce the parameter modeling bias. The existing non-parametric methods in the statistical methods are usually not applicable to software reliability data. In this paper we construct some non-parametric methods to estimate the failure intensity function of the NHPP model, taking the particularities of the software failure data into consideration.