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.     

A NOTE ON THE MULTIVARIATE LINEAR MODEL WITH CONSTRAINTS ON THE DEPENDENT VECTOR

It is shown that the least squares estimators of B and Σ in the multivariate linear model {E Y _i= X ₁ B , D ( Y _i) =Σ, 1 ≤ i ≤ n , Y ₁ Y _n uncorrelated} subject to the constraints Y _i M = X _i N are just the usual least squares estimators = ( X'X )-¹ X'Y and ΣC = 1/n( Y-X )( Y-X ) in the unconstrained model where Σ has full rank. Tests of hypotheses concerning B are discussed for situations in which each Y _i has a multivariate normal distribution, and examples of the applicability of the model reviewed.     

COMBINING INFORMATION FROM LINEAR MODELS ON POSSIBLY DIFFERENT SCALES

This paper examines the joint statistical analysis of M independent data sets, the jth of which satisfies the model λ_j Y_j=X_jB +ε_j, where the λ_j are unknown and the ε_i are normally distributed with a known correlation structure. The maximum likelihood equations, their asymptotic covariance matrix, and the likelihood ratio test of the hypothesis that the λ_js are all equal are derived. These results are applied to two examples.     

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.     

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.     

A ROBUST BAYES FACTOR FOR LINEAR MODELS

This paper proposes a new robust Bayes factor for comparing two linear models. The factor is based on a pseudo‐model for outliers and is more robust to outliers than the Bayes factor based on the variance‐inflation model for outliers. If an observation is considered an outlier for both models this new robust Bayes factor equals the Bayes factor calculated after removing the outlier. If an observation is considered an outlier for one model but not the other then this new robust Bayes factor equals the Bayes factor calculated without the observation, but a penalty is applied to the model considering the observation as an outlier. For moderate outliers where the variance‐inflation model is suitable, the two Bayes factors are similar. The new Bayes factor uses a single robustness parameter to describe a priori belief in the likelihood of outliers. Real and synthetic data illustrate the properties of the new robust Bayes factor and highlight the inferior properties of Bayes factors based on the variance‐inflation model for outliers.     

A NOTE ON THE MAXIMUM LIKELIHOOD ESTIMATION OF REGRESSION MODELS WITH FIRST ORDER MOVING AVERAGE ERRORS WITH ROOTS ON THE UNIT CIRCLE

This note gives an alternative derivation of Sargan and Bhargava' s results concerning the probability of observing on the unit circle a local maximum of the likelihood function of regression models with first order moving average errors. The note also discusses the relevance of these results for the computation of the exact maximum likelihood estimation of regression models with moving average errors.     

JAMES-STEIN RULE ESTIMATORS IN LINEAR REGRESSION MODELS WITH MULTIVARIATE-t DISTRIBUTED ERROR

This paper considers estimation of β in the regression model y =Xβ+μ, where the error components in μ have the jointly multivariate Student-t distribution. A family of James-Stein type estimators (characterised by nonstochastic scalars) is presented. Sufficient conditions involving only X are given, under which these estimators are better (with respect to the risk under a general quadratic loss function) than the usual minimum variance unbiased estimator (MVUE) of β. Approximate expressions for the bias, the risk, the mean square error matrix and the variance-covariance matrix for the estimators in this family are obtained. A necessary and sufficient condition for the dominance of this family over MVUE is also given.     

A NOTE ON ARMA PARAMETERISATION*

This paper considers the relationship between ARMA parameterisations of models for y(t) and Ay(t), where A is invertible and y(t) is a vector time series (t = 0,±1,…). An ARMA model for the transformed series Ay(t) may have fewer parameters than a model for y(t). This paper shows that such a saving is illusory because the apparently saved parameters are exactly balanced by the number of new parameters appearing in A.     

A NOTE ON THE APPLICATION OF THE KALMAN FILTER TO REGRESSION MODELS WITH SOME PARAMETERS VARYING OVER TIME AND OTHERS UNVARYING

The Kalman filter has been applied to estimation of the time-varying vector of regression parameters. I investigate the case where a portion of elements of the vector is invariant over time while others are varying as generated by the nonstationary, random walk model. Combined with the regression model it yields a state-space model in which observability holds but controllability does not. Under Grenan-der's condition on the exogenous variables I shall show that the estimate of the time-invariant portion is consistent, despite the seemingly unfavorable circumstances mentioned above, with the order equal to the reciprocal of sample size.     

ON THE CAUSALITY TEST IN TIME SERIES MODELS WITH HEAVY-TAILED DISTRIBUTION

ABSTRACT

In this article, we consider the problem of testing the Granger causality in stationary time series models with non-normal heavy-tailed distributions. We consider a normal mixture model to cover the heavy-tailed distribution, and propose a test statistic based on the partially adaptive estimator proposed by Phillips [1] Phillips, R.F. 1994. Partially Adaptive Estimation via a Normal Mixture. J. Econometics, 64: 123–144. [Crossref], [Web of Science ®] , [Google Scholar]. It is shown that the test statistic asymptotically follows a chi-squared distribution. Simulation results indicate that our test outperforms the conventional test based on the least squares estimator when the observations follow a heavy-tailed distribution.     

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.     

BOOTSTRAP TESTS FOR THE ERROR DISTRIBUTION IN LINEAR AND NONPARAMETRIC REGRESSION MODELS

In this paper we investigate several tests for the hypothesis of a parametric form of the error distribution in the common linear and non‐parametric regression model, which are based on empirical processes of residuals. It is well known that tests in this context are not asymptotically distribution‐free and the parametric bootstrap is applied to deal with this problem. The performance of the resulting bootstrap test is investigated from an asymptotic point of view and by means of a simulation study. The results demonstrate that even for moderate sample sizes the parametric bootstrap provides a reliable and easy accessible solution to the problem of goodness‐of‐fit testing of assumptions regarding the error distribution in linear and non‐parametric regression models.     

A mixture of the manova and gmanova models

A mixture of the MANOVA and GMANOVA models is presented. The expected value of the response matrix in this model is the sum of two matrix components. The first component represents the GMANOVA portion and the second component represents the MANOVA portion. Maximum likelihood estimators are derived for the parameters in this model, and goodness-of-fit tests are constructed for fuller models via the likelihood ration criterion. Finally, likelihood ration tests for general liinear hypotheses are developed and a numerical example is presented.     

A NOTE ON SAMPLING DESIGNS FOR RANDOM PROCESSES WITH NO QUADRATIC MEAN DERIVATIVE

Several authors have previously discussed the problem of obtaining asymptotically optimal design sequences for estimating the path of a stochastic process using intricate analytical techniques. In this note, an alternative treatment is provided for obtaining asymptotically optimal sampling designs for estimating the path of a second order stochastic process with known covariance function. A simple estimator is proposed which is asymptotically equivalent to the full‐fledged best linear unbiased estimator and the entire asymptotics are carried out through studying this estimator. The current approach lends an intuitive statistical perspective to the entire estimation problem.     

A REPARAMETERISATION METHOD FOR EMBEDDED MODELS

ABSTRACT

A reparameterisation procedure is investigated for embedded model problems. The procedure is given by solving differential equations determined by indeterminate forms of limit. Some properties are provided for the existence of an embedded model. Note that an embedded model may include another embedded model. We introduce the concept of embedded model of kth generation and discuss the use of one-by-one elimination procedure to construct graphs of embedded models. As examples, we derive embedded models for some distributions, to which existing method cannot be applied. Our method includes the method given by Cheng et al. [1] Cheng, R.C.H., Evans, B.E. and Iles, T.C. 1992. Embedded Models in Non-Linear Regression. J. R. Statist. Soc. B, 54: 877–888.  [Google Scholar] as a special case.     

A monte carlo study of the power of alternative tests under the generalized manova model

Monte Carlo simulations are performed for a broad range of conditions. These simulations indicate that the powers of alternative tests under the generalized MANOVA model for small samples differ significantly, if a large reduction of the number of polynomial parameters is applied. The results show that, if the response covariance matrix ∑ is known, the best alternative is to use ∑. If, however, ∑ is unknown, substitution of an identity matrix for ∑ is recommended. This alternative usually results in a test with more power than the test with the usual estimate of ∑ employing covariates or the test with an estimate of E obtained from another sample.     

A NOTE ON VARIANCE ESTIMATION FOR THE GENERALIZED REGRESSION PREDICTOR

The generalized regression (GREG) predictor is used for estimating a finite population total when the study variable is well‐related to the auxiliary variable. In 1997, Chaudhuri & Roy provided an optimal estimator for the variance of the GREG predictor within a class of non‐homogeneous quadratic estimators (H) under a certain superpopulation model M. They also found an inequality concerning the expected variances of the estimators of the variance of the GREG predictor belonging to the class H under the model M. This paper shows that the derivation of the optimal estimator and relevant inequality, presented by Chaudhuri & Roy, are incorrect.