Optimal designs for generalized linear models

Summary This paper solves some D-optimal design problems for certain Generalized Linear Models where the mean depends on two parameters and two explanatory variables. In all of the cases considered the support point of the optimal designs are found to be independent of the unknown parameters. While in some cases the optimal design measures are given by two points with equal weights, in others the support is given by three point with weights depending on the unknown parameters, hence the designs are locally optimal in general. Empirical results on the efficiency of the locally optimal designs are also given. Some of the designs found can also be used for planning D-optimal experiments for the normal linear model, where the mean must be positive. This research was carried out in part at University College, London as an M.Sc. project. Thanks are due to Prof. I. Ford (University of Glasgow) and Prof. A. Giovagnoli (University of Perugia) for their valuable suggestions and critical observations.     

D-optimal design for a quadratic log contrast model for experiments with mixtures

A practical method is suggested for solving complicated D-optimal design problems analytically. Using this method the author has solved the problem for a quadratic log contrast model for experiments with mixtures introduced by J. Aitchison and J. Bacon-Shone. It is found that for a symmetric subspace of the finite dimensional simplex, the vertices and the centroid of this subspace are the only possible support points for a D-optimal design. The weights that must be assigned to these support points contain irrational numbers and are constrained by a system of three simultaneous linear equations, except for the special cases of 1- and 2-dimensional simplexes where the situation is much simpler. Numerical values for the solution are given up to the 19-dimensional simplex     

Modified d-optimal sequential procedure in allocation problem

When an experimenter wishes to compare t treatments with M experimental units, one of the first problems he faces is how to allocate N experimental units into t treatments. When no pre treat merit information about the experimental units is available, "randomization" is the widely accepted guiding principle to deal with the allocation problem But pre treat merit information usually is available, although it is seldom fully used for allocation purposes. Recently, Harville considered the allocation problem under a covariance model. He suggested a D-optimal sequential procedure that may be used to construct nearly D-optimal allocations. However, Harville's sequential procedure requires constructing a D-optimal initial allocation at the first stages and that may be computationally unfeasible in some real situations, Such construction is not needed for a new sequential.     

Optimal designs for minimising covariances among parameter estimators in a linear model

We construct approximate optimal designs for minimising absolute covariances between least‐squares estimators of the parameters (or linear functions of the parameters) of a linear model, thereby rendering relevant parameter estimators approximately uncorrelated with each other. In particular, we consider first the case of the covariance between two linear combinations. We also consider the case of two such covariances. For this we first set up a compound optimisation problem which we transform to one of maximising two functions of the design weights simultaneously. The approaches are formulated for a general regression model and are explored through some examples including one practical problem arising in chemistry.     

Optimal designs in random coefficient cubic regression models

In this article we investigate the problem of ascertaining A- and D-optimal designs in a cubic regression model with random coefficients. Our interest lies in estimation of all the parameters or in only those except the intercept term. Assuming the variance ratios to be known, we tabulate D-optimal designs for various combinations of the variance ratios. A-optimality does not pose any new problem in the random coefficients situation.     

Bayesian D-optimal design for logistic regression model with exponential distribution for random intercept

ABSTRACT

Nowadays, generalized linear models have many applications. Some of these models which have more applications in the real world are the models with random effects; that is, some of the unknown parameters are considered random variables. In this article, this situation is considered in logistic regression models with a random intercept having exponential distribution. The aim is to obtain the Bayesian D-optimal design; thus, the method is to maximize the Bayesian D-optimal criterion. For the model was considered here, this criterion is a function of the quasi-information matrix that depends on the unknown parameters of the model. In the Bayesian D-optimal criterion, the expectation is acquired in respect of the prior distributions that are considered for the unknown parameters. Thus, it will only be a function of experimental settings (support points) and their weights. The prior distribution of the fixed parameters is considered uniform and normal. The Bayesian D-optimal design is finally calculated numerically by R3.1.1 software.     

A variable-neighbourhood search algorithm for finding optimal run orders in the presence of serial correlation

The responses obtained from response surface designs that are run sequentially often exhibit serial correlation or time trends. The order in which the runs of the design are performed then has an impact on the precision of the parameter estimators. This article proposes the use of a variable-neighbourhood search algorithm to compute run orders that guarantee a precise estimation of the effects of the experimental factors. The importance of using good run orders is demonstrated by seeking D-optimal run orders for a central composite design in the presence of an AR(1) autocorrelation pattern.     

On invariance and randomization in factorial designs with applications to D-optimal main effect designs of the symmetrical factorial

Under the setting of the columnwise orthogonal polynomial model in the context of the general factorial it is shown that (i) the determinant of the information matrix of a design relative to an admissible vector of effects is invariant under a permutation of levels; (ii) the unbiased estimation of a linear function of an admissible vector of effects can be obtained under equal probability randomization. These results extend the work on invariance and randomization carried out under the more restrictive assumption of the orthonormal polynomial model by Srivastava, Raktoe and Pesotan (1976) and Pesotan and Raktoe (1981). Moreover, the problem of the construction of D-optimal main effect designs in the symmetrical factorial is reduced to a study of a special class of (0,1)-matrices using the Helmat matrix model. Using this class of (0,1)-matrices and the determinant invariance result, some classes of D-optimal main effect designs of the s² and s³ factorial respectively are presented.     

Optimal and efficient designs for 2-parameter nonlinear models

By Carathéodory's theorem, for a k-parameter nonlinear model, the minimum number of support points for any D-optimal design is between k and k(k+1)/2. Characterizing classes of models for which a D-optimal design sits on exactly k support points is of great theoretical interest. By utilizing the equivalence theorem, we identify classes of 2-parameter nonlinear models for which a D-optimal design is precisely supported on 2 points. We also introduce the theory of maximum principle from differential equations into the design area and obtain some results on characterizing the minimally supported nonlinear designs. Examples are given to demonstrate our results. Designs with minimum number of support points may not always be suitable in practice. To alleviate this problem, we utilize some geometric and analytical methods to obtain some efficient designs which provide more opportunity for the model checking and prevent biases due to mis-specified initial parameters.     

Minimax designs for 2k factorial experiments for generalized linear models

ABSTRACT

Formulas for A- and C-optimal allocations for binary factorial experiments in the context of generalized linear models are derived. Since the optimal allocations depend on GLM weights, which often are unknown, a minimax strategy is considered. This is shown to be simple to apply to factorial experiments. Efficiency is used to evaluate the resulting design. In some cases, the minimax design equals the optimal design. For other cases no general conclusion can be drawn. An example of a two-factor logit model suggests that the minimax design performs well, and often better than a uniform allocation.     

Application of genetic algorithms to the construction of exact D-optimal designs

This paper studies the application of genetic algorithms to the construction of exact D-optimal experimental designs. The concept of genetic algorithms is introduced in the general context of the problem of finding optimal designs. The algorithm is then applied specifically to finding exact D-optimal designs for three different types of model. The performance of genetic algorithms is compared with that of the modified Fedorov algorithm in terms of computing time and relative efficiency. Finally, potential applications of genetic algorithms to other optimality criteria and to other types of model are discussed, along with some open problems for possible future research.     

Considerations on group-wise identical designs for linear mixed models

We consider a general class of mixed models, where the individual parameter vector is composed of a linear function of the population parameter vector plus an individual random effects vector. The linear function can vary for the different individuals. We show that the search for optimal designs for the estimation of the population parameter vector can be restricted to the class of group-wise identical designs, i.e., for each of the groups defined by the different linear functions only one individual elementary design has to be optimized. A way to apply the result to non-linear mixed models is described.     

D-optimal designs with minimal and nearly minimal number of units

D-optimal designs are identified in classes of connected block designs with fixed block size when the number of experimental units is one or two more than the minimal number required for the design to be connected. An application of one of these results is made to identify D-optimal designs in a class of minimally connected row-column designs. Graph-theoretic methods are employed to arrive at the optimality results.     

Experimental design for four mixture components with process variables

The experimental design to model the response of a mixture in four components in the presence of process variables is considered. Two different blocks of blends that are orthogonal for linear or quadratic blending are D-optimized. The two orthogonal blocks of blends are generalized and D-optimized in some cases (and possibly Doptimized in others) to deal with restrictions on the blending component proportions. The pair of orthogonal D-optimal blocks of blends can be used with an arbitrary number of process variables, and requires a reduced number of observations.     

On biased linear estimators in models with arbitrary rank

In this paper we define a class of biased linear estimators for the unknown parameters in linear models with arbitrary rank. The feature of our approach is to reduce the estimation problem in arbitrary rank models to the one in full-rank models. Some important properties are discussed. As special cases of our class, we extend to deficient-rank models six known biased linear estimators.     

Remainder modified systematic sampling in the presence of linear trend

If the population size is not a multiple of the sample size, then the usual linear systematic sampling design is unattractive, since the sample size obtained will either vary, or be constant and different to the required sample size. Only a few modified systematic sampling designs are known to deal with this problem and in the presence of linear trend, most of these designs do not provide favorable results. In this paper, a modified systematic sampling design, known as remainder modified systematic sampling (RMSS), is introduced. There are seven cases of RMSS and the results in this paper suggest that the proposed design is favorable, regardless of each case, while providing linear trend-free sampling results for three of the seven cases. To obtain linear trend-free sampling for the other cases and thus improve results, an end corrections estimator is constructed.     

Ratios of linear combinations in the general linear model

We consider the problem of estimating. the ratio of two linear combinations of thevector of parameters in the general linear model. The nonexistence of an unbiased estimator under normal errorsisdiscussed. Properties of an often used estimator, the maximum likelihood estimator under normal errors, are presented, This is done both for fixed sample size and asymptotically, in the

presence of normal and non-normal errors.     

Extended linear empirical bayes estimation

In this paper, the linear empirical Bayes estimation method, which is based on approximation of the Bayes estimator by a linear function, is generalized to an extended linear empirical Bayes estimation technique which represents the Bayes estimator by a series of algebraic polynomials. The extended linear empirical Bayes estimators are elaborated in the case of a location or a scale parameter. The theory is illustrated by examples of its application to the normal distribution with a location parameter and the gamma distribution with a scale parameter. The linear and the extended linear empirical Bayes estimators are constructed in these two cases and, then, studied numerically via Monte Carlo simulations. The simulations show that the extended linear empirical Bayes estimators have better convergence rates than the traditional linear empirical Bayes estimators.     

Atypical observations in linear models

The problem of ‘atypical data point’ in the estimation of model parameters and its effect on prediction are discussed. A cross-validity procedure is then proposed to accommodate the unusual observations in the estimation and thereby to improve the prediction of future data points. Each atypical point, weighted according to cross-validitory procedure, is used in the estimation of model parameters.     

On the potential in the estimation of linear functions in regression

New aspects of potential in Cook's Influence measure for linear combinations are explored. It is shown that this potential can be considered as a case influence measure in the scatter of estimated combinations. The potential is related to precise estimation directions and multicollinearity concepts; It Is also used as a basis for selection of new cases.