EMPIRICAL LIKELIHOOD‐BASED INFERENCES FOR PARTIALLY LINEAR MODELS WITH MISSING COVARIATES

This paper considers statistical inference for partially linear models Y = X ^? β +ν(Z) +? when the linear covariate X is missing with missing probability π depending upon (Y, Z). We propose empirical likelihood‐based statistics to construct confidence regions for β and ν(z). The resulting empirical likelihood ratio statistics are shown to be asymptotically chi‐squared‐distributed. The finite‐sample performance of the proposed statistics is assessed by simulation experiments. The proposed methods are applied to a dataset from an AIDS clinical trial.     

INCOMPLETE DATA IN GENERALIZED LINEAR MODELS WITH CONTINUOUS COVARIATES

This paper proposes a method for estimating the parameters in a generalized linear model with missing covariates. The missing covariates are assumed to come from a continuous distribution, and are assumed to be missing at random. In particular, Gaussian quadrature methods are used on the E-step of the EM algorithm, leading to an approximate EM algorithm. The parameters are then estimated using the weighted EM procedure given in Ibrahim (1990). This approximate EM procedure leads to approximate maximum likelihood estimates, whose standard errors and asymptotic properties are given. The proposed procedure is illustrated on a data set.     

PARAMETRIC FRACTIONAL IMPUTATION FOR NON‐IGNORABLE CATEGORICAL MISSING DATA WITH FOLLOW‐UP

Incomplete data subject to non‐ignorable non‐response are often encountered in practice and have a non‐identifiability problem. A follow‐up sample is randomly selected from the set of non‐respondents to avoid the non‐identifiability problem and get complete responses. Glynn, Laird, & Rubin analyzed non‐ignorable missing data with a follow‐up sample under a pattern mixture model. In this article, maximum likelihood estimation of parameters of the categorical missing data is considered with a follow‐up sample under a selection model. To estimate the parameters with non‐ignorable missing data, the EM algorithm with weighting, proposed by Ibrahim, is used. That is, in the E‐step, the weighted mean is calculated using the fractional weights for imputed data. Variances are estimated using the approximated jacknife method. Simulation results are presented to compare the proposed method with previously presented methods.     

ESTIMATION FOR THE GENERAL SAMPLE SELECTION MODELS

Consider a general regression model with an arbitrary and unknown link function and a stochastic selection variable that determines whether the outcome variable is observable or missing. The paper proposes U-statistics that are based on kernel functions as estimators for the directions of the parameter vectors in the link function and the selection equation, and shows that these estimators are consistent and asymptotically normal.     

MULTIPLE-RECORD SYSTEMS ESTIMATION USING LATENT CLASS MODELS

Capture–recapture methods (also referred to as 'multiple-record systems') have been widely used in enumerating human populations in the fields of epidemiology and public health. In this article, we introduce latent class models into multiple-record systems to account for unobserved heterogeneity in the population. Two approaches, the full and the conditional likelihood, are proposed to estimate the unknown population abundance. We also suggest rules to diagnose identifiability of the proposed latent class models. The methodologies are illustrated by two real examples: the first is to count the undercount of homelessness in the Adelaide central business district, and the second concerns the incidence of diabetes in a small Italian town.     

ON QUASI-LIKELIHOOD METHODS AND ESTIMATION FOR BUNCHING PROCESSES AND HETEROSCEDASTIC REGRESSION MODELS

This paper is concerned with quasi-likelihood estimation methods which provide benchmarks in efficiency against which other methods may be as sessed. We discuss general principles and illustrate them by the estimation of parameters for branching processes with immigration and heteroscedastic regression models.     

EFFICIENT ESTIMATION IN REGRESSION FRAILTY OR TRANSFORMATION MODELS BASED ON AN ALGORITHM

This paper considers a regression frailty or transformation model in which the structural parameter is the vector of regression coefficients and the nuisance parameter is a vector of arbitrarily high dimension. It proposes jointly (implicitly) defined parameter estimators which have been proved to be consistent and asymptotically efficient, and develops an algorithmic procedure that provides these estimators. The behaviour of the algorithm is illustrated by analysing simulated and real data.     

EXACT P‐VALUES FOR DISCRETE MODELS OBTAINED BY ESTIMATION AND MAXIMIZATION

In constructing exact tests from discrete data, one must deal with the possible dependence of the P‐value on nuisance parameter(s) ψ as well as the discreteness of the sample space. A classical but heavy‐handed approach is to maximize over ψ. We prove what has previously been understood informally, namely that maximization produces the unique and smallest possible P‐value subject to the ordering induced by the underlying test statistic and test validity. On the other hand, allowing for the worst case will be more attractive when the P‐value is less dependent on ψ. We investigate the extent to which estimating ψ under the null reduces this dependence. An approach somewhere between full maximization and estimation is partial maximization, with appropriate penalty, as introduced by Berger & Boos (1994, P values maximized over a confidence set for the nuisance parameter. J. Amer. Statist. Assoc. 89 , 1012–1016). It is argued that estimation followed by maximization is an attractive, but computationally more demanding, alternative to partial maximization. We illustrate the ideas on a range of low‐dimensional but important examples for which the alternative methods can be investigated completely numerically.     

ROBUST ESTIMATION AND HYPOTHESIS TESTS FOR FIRST-ORDER THRESHOLD AUTOREGRESSIVE MODELS

Results of Petrucelli & Woolford (1984) for a first-order threshold autoregressive model are considered from a robust point of view. Robust estimators of the threshold parameters of the model are obtained and their asymptotic normality is proved. Testing the equality of the threshold parameters is considered using the robust analogues of Wald and score test statistics. Limiting distributions of these statistics are given under both null and alternative hypotheses.     

NON-PARAMETRIC ESTIMATION OF DIRECTION IN SINGLE-INDEX MODELS WITH CATEGORICAL PREDICTORS

This paper proposes a general dimension‐reduction method targeting the partial central subspace recently introduced by Chiaromonte, Cook & Li. The dependence need not be confined to particular conditional moments, nor are restrictions placed on the predictors that are necessary for methods like partial sliced inverse regression. The paper focuses on a partially linear single‐index model. However, the underlying idea is applicable more generally. Illustrative examples are presented.     

MAXIMUM LIKELIHOOD ESTIMATION FOR A POISSON RATE PARAMETER WITH MISCLASSIFIED COUNTS

This paper proposes a Poisson‐based model that uses both error‐free data and error‐prone data subject to misclassification in the form of false‐negative and false‐positive counts. It derives maximum likelihood estimators (MLEs) for the Poisson rate parameter and the two misclassification parameters — the false‐negative parameter and the false‐positive parameter. It also derives expressions for the information matrix and the asymptotic variances of the MLE for the rate parameter, the MLE for the false‐positive parameter, and the MLE for the false‐negative parameter. Using these expressions the paper analyses the value of the fallible data. It studies characteristics of the new double‐sampling rate estimator via a simulation experiment and applies the new MLE estimators and confidence intervals to a real dataset.     

A SEMIPARAMETRIC REGRESSION MODEL WITH MISSING COVARIATES IN CONTINUOUS-TIME CAPTURE-RECAPTURE STUDIES

Covariate data were missing when a semiparametric regression model was used to study bird abundance in the Mai Po Sanctuary, Hong Kong. This paper proposes an EM‐type algorithm to estimate the regression parameters for that study. Analytical calculation of the expectation in the EM method is difficult, or even impossible, especially when missing covariates are continuous. A Monte Carlo method is used in the EM algorithm to ease the calculation complexity. Asymptotic variances of the parameter estimates are also derived. Properties of the proposed estimators are assessed through numerical simulations and a real example.     

OPTIMAL QUADRATIC UNBIASED ESTIMATION FOR MODELS WITH LINEAR TOEPLITZ COVARIANCE STRUCTURE

This paper deals with the problem of quadratic unbiased estimation for models with linear Toeplitz covariance structure. These serial covariance models are very useful to modelize time or spatial correlations by means of linear models. Optimality and local optimality is examined in different ways. For the nested Toeplitz models, it is shown that there does not exist a Uniformly Minimum Variance Quadratic Unbiased Estimator for at least one linear combination of covariance parameters. Moreover, empirical unbiased estimators are identified as Locally Minimum Variance Quadratic Unbiased Estimators for a particular choice on covariance parameters corresponding to the case where the covariance matrix of the observed random vector is proportional to the identity matrix. The complete Toeplitz-circulant model is also studied. For this model, the existence of a Uniformly Minimum Variance Quadratic Unbiased Estimator for each covariance parameter is proved.     

ASYMPTOTIC TESTS FOR GROWTH CURVE MODELS WITH AUTOREGRESSIVE ERRORS

This paper presents the limit distribution (as the number of time points increase) for the score vector of a growth curve model assuming both stationary and explosive autoregressive (A.R.) errors. Limit distributions of the score statistic and the likelihood-ratio statistic for testing composite hypotheses about the regression parameters of several growth curves, when the autocorrelation parameters are treated as nuisance parameters, are presented.     

RATES OF CONVERGENCE IN SEMI-PARAMETRIC MODELLING OF LONGITUDINAL DATA

We consider the problem of semi-parametric regression modelling when the data consist of a collection of short time series for which measurements within series are correlated. The objective is to estimate a regression function of the form E[Y(t) | x] =x'ß+μ(t), where μ(.) is an arbitrary, smooth function of time t, and x is a vector of explanatory variables which may or may not vary with t. For the non-parametric part of the estimation we use a kernel estimator with fixed bandwidth h. When h is chosen without reference to the data we give exact expressions for the bias and variance of the estimators for β and μ(t) and an asymptotic analysis of the case in which the number of series tends to infinity whilst the number of measurements per series is held fixed. We also report the results of a small-scale simulation study to indicate the extent to which the theoretical results continue to hold when h is chosen by a data-based cross-validation method.     

GRAPHICAL SENSITIVITY ANALYSIS WITH DIFFERENT METHODS OF IMPUTATION FOR A TRIAL WITH PROBABLE NON‐IGNORABLE MISSING DATA

Graphical sensitivity analyses have recently been recommended for clinical trials with non‐ignorable missing outcome. We demonstrate an adaptation of this methodology for a continuous outcome of a trial of three cognitive‐behavioural therapies for mild depression in primary care, in which one arm had unexpectedly high levels of missing data. Fixed‐value and multiple imputations from a normal distribution (assuming either varying mean and fixed standard deviation, or fixed mean and varying standard deviation) were used to obtain contour plots of the contrast estimates with their P‐values superimposed, their confidence intervals, and the root mean square errors. Imputation was based either on the outcome value alone, or on change from baseline. The plots showed fixed‐value imputation to be more sensitive than imputing from a normal distribution, but the normally distributed imputations were subject to sampling noise. The contours of the sensitivity plots were close to linear in appearance, with the slope approximately equal to the ratio of the proportions of subjects with missing data in each trial arm.     

A NOTE ON ESTIMATING LINEAR TREND IN A REGRESSION MODEL WITH SERIALLY CORRELATED ERROR COMPONENTS

ABSTRACT

We consider a linear trend regression model when the disturbances follow a serially correlated one-way error component model. In this model, we investigate the performance of the Ordinary Least Squares Esitmator (OLSE), First Difference Estimator (FDE), Generalized Least Squares Estimator (GLSE) and the Cochrane-Orcutt-Transformation Estimator (COTE) of slope coefficient in terms of efficiency. The main findings are as follows: (1) when the autocorrelation is close to unity, then the FDE is approximately the GLSE; (2) the OLSE is better than the COTE; and (3) when the value of the autocorrelation is kept constant and T → ∞, the OLSE, COTE and GLSE are asymptotically equivalent whereas the FDE is worse than the other estimators in terms of efficiency.     

Nonlinear mixed‐effect models with nonignorably missing covariates

Nonlinear mixed‐effect models are often used in the analysis of longitudinal data. However, it sometimes happens that missing values for some of the model covariates are not purely random. Motivated by an application to HTV viral dynamics, where this situation occurs, the author considers likelihood inference for this type of problem. His approach involves a Monte Carlo EM algorithm, along with a Gibbs sampler and rejection/importance sampling methods. A concrete application is provided.