ON CONFIDENCE REGIONS OF EMBEDDED MODELS IN REGULAR PARAMETRIC FAMILIES (A GEOMETRIC APPROACH)

Regular parametric families are commonly encountered in statistical problems (e.g. Cox & Hinkley, 1974). In this paper, we propose a differential geometric framework for the embedded models in these families. Our framework may be regarded as an extension of that presented by Bates & Watts (1980) for nonlinear regression models. As an application, we use this geometric framework to derive three kinds of improved approximate confidence regions for the parameter and parameter subsets in terms of curvatures. The results obtained by Hamilton et al. (1982) and Hamilton (1986) are extended to embedded models in regular parametric families.     

A METHOD FOR ESTIMATING PARAMETER IN NONNEGATIVE MA(1) MODELS

ABSTRACT

A method for estimating parameter in nonnegative MA(1) models is proposed and investigated in the paper. The method also gives nontrivial confidence sets on confidence level 1. Small sample properties of new estimator are demonstrated in a simulation study.     

SKEWNESS FOR PARAMETERS IN GENERALIZED LINEAR MODELS

In this paper we give an asymptotic formula of order n ^?1/2, where n is the sample size, for the skewness of the distribution of the maximum likelihood estimates of the linear parameters in generalized linear models. The formula is given in matrix notation and is very suitable for computer implementation. Several special cases are discussed. We also give asymptotic formulae for the skewness of the distribution of the maximum likelihood estimates of the dispersion and precision parameters.     

A SEMIPARAMETRIC HIERARCHICAL METHOD FOR A REGRESSION MODEL WITH AN INTERVAL-CENSORED COVARIATE

A Bayesian framework is proposed for analysing regression models in which one of the covariates is interval‐censored. Such a situation was encountered in an AIDS clinical trial in which the goal was to examine the association between delays in initiating a new treatment after Indinavir failure and the subsequent viral load level of patients at the time of enrolment into the new treatment. The new method uses a mixture of Dirichlet processes allowing all the components in the model to be specified parametrically, except for the distribution of the interval‐censored covariate, which is treated non‐parametrically. The paper explains the proposed method for the linear regression model in detail. The performance of the method is assessed by simulations and illustrated using the AIDS clinical trial.     

A NOTE ON MULTIVARIATE LOGISTIC MODELS FOR CONTINGENCY TABLES

The log-linear model is a tool widely accepted for modelling discrete data given in a contingency table. Although its parameters reflect the interaction structure in the joint distribution of all variables, it does not give information about structures appearing in the margins of the table. This is in contrast to multivariate logistic parameters, recently introduced by Glonek & McCullagh (1995), which have as parameters the highest order log odds ratios derived from the joint table and from each marginal table. Glonek & McCullagh give the link between the cell probabilities and the multivariate logistic parameters, in an algebraic fashion. The present paper focuses on this link, showing that it is derived by general parameter transformations in exponential families. In particular, the connection between the natural, the expectation and the mixed parameterization in exponential families (Barndorff-Nielsen, 1978) is used; this also yields the derivatives of the likelihood equation and shows properties of the Fisher matrix. The paper emphasises the analysis of independence hypotheses in margins of a contingency table.     

DECOMPOSITIONS FOR CONDITIONAL SYMMETRY MODEL INTO PALINDROMIC SYMMETRY AND MODIFIED MARGINAL HOMOGENEITY MODELS

For square contingency tables with ordered categories, the conditional symmetry model is decomposed into the palindromic symmetry and the modified marginal homogeneity models, and moreover into the generalized palindromic symmetry model and two other models. The data on unaided vision first analysed by Stuart (1953, 1955) is analysed again by using the decompositions.     

ASYMPTOTIC PROPERTIES OF RESTRICTED MAXIMUM LIKELIHOOD (REML) ESTIMATES FOR HIERARCHICAL MIXED LINEAR MODELS

This paper explores the asymptotic distribution of the restricted maximum likelihood estimator of the variance components in a general mixed model. Restricting attention to hierarchical models, central limit theorems are obtained using elementary arguments with only mild conditions on the covariates in the fixed part of the model and without having to assume that the data are either normally or spherically symmetrically distributed. Further, the REML and maximum likelihood estimators are shown to be asymptotically equivalent in this general framework, and the asymptotic distribution of the weighted least squares estimator (based on the REML estimator) of the fixed effect parameters is derived.     

A FULLY NONPARAMETRIC TEST FOR ONE-WAY LAYOUTS

The use of a vector of sign statistics as the basis of a nonparametric test for equality of distributions in one‐way layouts is considered. An important feature of this test is its ability to detect a broad range of alternatives, including scale and shape differences. In this scenario, the data consist of several independent measurements on each treatment or subject. Presented here are a finite sample and asymptotic distribution theory for the test statistics and a discussion of a follow‐up clustering procedure based on a mixture model approximation. Finally, the methods are illustrated using two examples from the literature.     

AN ITERATIVE METHOD FOR ESTIMATION OF PARAMETERS IN A CLUSTERED SAMPLING MODEL

This paper considers an iterative method for obtaining maximum likelihood estimates for a contingency table derived from a clustered sampling model. Comparisons are made with other methods proposed in the literature.     

A NOTE ON MULTIVARIATE LINEAR MODELS WITH NON-ADDITIVITY

Tukey's non-additivity test in an analysis of variance model is extended to a multivariate linear model with covariates. If non-additivity is found to exist, a Wilks's Lambda test for the dimensionality of the matrix of the non-additivity parameters is derived and the Lambda criterion is then factorized into two independent test criteria to test meaningful hypotheses concerning the multivariate model.     

SMOOTH TRANSITION AUTOREGRESSIVE MODELS — A SURVEY OF RECENT DEVELOPMENTS

This paper surveys recent developments related to the smooth transition autoregressive (STAR) time series model and several of its variants. We put emphasis on new methods for testing for STAR nonlinearity, model evaluation, and forecasting. Several useful extensions of the basic STAR model, which concern multiple regimes, time-varying non-linear properties, and models for vector time series, are also reviewed.     

A METHOD FOR DETERMINING THE ASYMPTOTIC EFFICIENCY OF SOME SEQUENTIAL PROBABILITY RATIO TESTS

This paper gives a method for decomposing many sequential probability ratio tests into smaller independent components called “modules”. A function of some characteristics of modules can be used to determine the asymptotically most efficient of a set of statistical tests in which a, the probability of type I error equals β, the probability of type II error. The same test is seen also to give the asymptotically most efficient of the corresponding set of tests in which a is not equal to β. The “module” method is used to give an explanation for the super-efficiency of the play-the-winner and play-the-loser rules in two-sample binomial sampling. An example showing how complex cases can be analysed numerically using this method is also given.     

A SEMI-MARKOV MODEL FOR EAR INFECTION

A recent study of the prevalence of middle ear infection among Australian aborigines reports, for different age groups, the number currently infected, the number currently not infected and the number who are currently not infected but have scarring of the ear drum showing evidence of a previous infection. Incidence and recovery are modelled by a semi-Markov model in which incidence of the infection is allowed to be nonstationary.     

A new perspective in functional EIV linear model: Part I

Simple linear regression in the functional errors-in-variables (EIV) model is revisited from a different perspective, where the problem is addressed by using the small-sigma model instead of large sample theory. A general analysis is developed to study the slope’s estimator that minimizes a family of objective functions, of which the least-squares fit and the maximum likelihood estimator are minimizers of such special functions. General formulas for the higher-order terms of the bias, the variance, and the mean square error are derived. Accordingly, two efficient estimators are proposed after implementing the pre- and the post-bias elimination techniques. Numerical tests confirm the superiority of the proposed estimators over others.     

ESTIMATION OF QUANTILES FOR A FRÉCHET-TYPE DISTRIBUTION

Estimation of high quantiles of a distribution in the domain of attraction of the Fréchet distribution is based on the extremal distribution of the k largest order statistics. The problem is treated by a local maximum likelihood method on a three parameter model. The estimators are shown to be asymptotically consistent for the whole range of the tail index parameter.     

SIMULATIONS AND DERIVED APPROXIMATIONS FOR THE MEANS AND STANDARD DEVIATIONS OF THE CHARACTERISTIC ROOTS OF A WISHART MATRIX

Monte Carlo simulations were done to estimate the means and standard deviations of the characteristic roots of a Wishart matrix which can be used in computing tests of hypotheses concerning multiplicative terms in balanced linear-bilinear (multiplicative) models for an m × n table of data. In this report we extend the previous results (Mandel, 1971; Cornelius, 1980) to r ≤ 199, c ≤ 149 or r ≤ 149, c ≤ 199, where r and c are row and column degrees of freedom, respectively, of the two-way array of residuals (with total degrees of freedom rc) after fitting the linear effects. For 187 combinations of r and c at intervals over this domain, we used 5000 simulations to estimate expectations and standard deviations of the Wishart roots. Using weighted linear regression variable selection techniques, symmetric functions of r and c were obtained for approximating the simulated means and standard deviations. Use of these approximating functions will avoid the need for reference to tables for input to computer programs which require these values for tests of significance of sequentially fitted terms in the analyses of balanced linear-bilinear models.     

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.     

A special property of the expected log likelihooda special property of the expected log likelihood

It is shown that linear transformations of the logarithm are the only functions of the likelihood whose expected values discriminate between correct and incorrect likelihoods by a simple ordering property, assuming the correct probability density function is continuous. Also, an extension of this result is given for the predictive densities considered by Akaike.     

Non-nested linear models: A conditional confidence approach

The comparison of nested linear models with normal error is well standardized in the common procedures of the analysis of variance. This article considers the comparison of two non-nested linear models that have the same parameter dimension; the comparison is made on the assumption that the true mean lies somewhere in the linear span of the two models. The analysis leads to a precision-based conditional confidence interval for the unsigned angular direction of the true mean, and this in turn provides a confidence assessment of the two directions that correspond to the two models being compared. The confidence interval is an approximate conditional interval (given the distance of the estimate from the intersection of the hypotheses), and its length as a fraction of π indicates the precision of the confidence procedure. The method provides a conditional-inference alternative to a confidence interval available by Creasy-Fieller analysis.     

A Comparison of Analysis of Covariate-Adjusted Residuals and Analysis of Covariance

Various methods to control the influence of a covariate on a response variable are compared. These methods are ANOVA with or without homogeneity of variances (HOV) of errors and Kruskal–Wallis (K–W) tests on (covariate-adjusted) residuals and analysis of covariance (ANCOVA). Covariate-adjusted residuals are obtained from the overall regression line fit to the entire data set ignoring the treatment levels or factors. It is demonstrated that the methods on covariate-adjusted residuals are only appropriate when the regression lines are parallel and covariate means are equal for all treatments. Empirical size and power performance of the methods are compared by extensive Monte Carlo simulations. We manipulated the conditions such as assumptions of normality and HOV, sample size, and clustering of the covariates. The parametric methods on residuals and ANCOVA exhibited similar size and power when error terms have symmetric distributions with variances having the same functional form for each treatment, and covariates have uniform distributions within the same interval for each treatment. In such cases, parametric tests have higher power compared to the K–W test on residuals. When error terms have asymmetric distributions or have variances that are heterogeneous with different functional forms for each treatment, the tests are liberal with K–W test having higher power than others. The methods on covariate-adjusted residuals are severely affected by the clustering of the covariates relative to the treatment factors when covariate means are very different for treatments. For data clusters, ANCOVA method exhibits the appropriate level. However, such a clustering might suggest dependence between the covariates and the treatment factors, so makes ANCOVA less reliable as well.