Optimal prediction designs in finite discrete spectrum linear regression models

In the paper, we solve the n-point optimal prediction design problem for the simplest nontrivial finite discrete spectrum linear regression models with correlated observations. We show that for all the models in consideration, there exists an optimal prediction design supported on at most three distinct points, which can be computed using one-dimensional optimization. In some cases, an optimal prediction design can be found explicitly.     

Optimal and robust designs for trigonometric regression models

This article presents discussions on the optimal and robust designs for trigonometric regression models under different optimality criteria. First, we investigate the classical Q-optimal designs for estimating the response function in a full trigonometric regression model with a given order. The equivalencies of Q-, A-, and G-optimal designs for trigonometric regression in general are also articulated. Second, we study minimax designs and their implementation in the case of trigonometric approximation under Q-, A-, and D-optimality. Then, We indicate the existence of the symmetric designs that are D-optimal minimax designs for general trigonometric regression models, and prove the existence of the symmetric designs that are Q- or A-optimal minimax designs for two particular trigonometric regression models under certain conditions.     

Optimal investment decisions when time-horizon is uncertain

Many investors do not know with certainty when their portfolio will be liquidated. Should their portfolio selection be influenced by the uncertainty of exit time? In order to answer this question, we consider a suitable extension of the familiar optimal investment problem of Merton [Merton, R.C., 1971. Optimal consumption and portfolio rules in a continuous-time model. Journal of Economic Theory 3, 373–413], where we allow the conditional distribution function of an agent’s time-horizon to be stochastic and correlated to returns on risky securities. In contrast to existing literature, which has focused on an independent time-horizon, we show that the portfolio decision is affected.     

Optimal designs for linear mixture models

Summary In a recent paper S nee and M arquardt [8] considered designs for linear mixture models, where the components are subject to individual lower and/or upper bounds. When the number of components is large their algorithm XVERT yields designs far too extensive for practical purposes.
The purpose of this paper is to describe a numerical procedure resulting in a design of fixed size N , which is approximately D -optimal, and where the components may be subject to linear constraints (f.e. upper or lower bounds). The proposed method is more generally applicable for models linear in the independent variables and the parameters and the convex hull of the experimental region is a polyhedron whose vertices are known.     

Optimal block designs for diallel crosses

Dey and Midha (Biometrika 83(2):484–489, 1996) constructed optimal block designs for complete diallel cross experiment using triangular partially balanced incomplete block designs with two associate classes. They listed optimal block designs for the lines in the range from 5 ≤ v ≤ 10. In this paper, we are also proposing additional optimal block designs for complete diallel cross experiment using two associate class partially balanced block designs for some additional values of v. Our method yields designs for proper and non-proper settings for complete diallel cross experiments. The proper and non proper designs are optimal in the sense of Kempthorne (Genetics 41:451–459, 1956) and non-proper designs are universally optimal in the sense of Kiefer (A survey of statistical design and linear models, North Holland, Amsterdam, 1975). The list of practically important designs is also given.     

Optimal designs for a mixed interference model

This paper generalizes Kunert and Martin’s (Ann Stat 28:1728–1742, 2000) method for finding optimal designs under a fixed interference model, to find optimal designs under a mixed interference model. The results are based on the properties of information matrices in fixed and mixed models given in Markiewicz (J Stat Plan Inference 59:127–137, 1997). The method is applied to find a design which is optimal for any given variances of random neighbor effects. Research partially supported by the KBN Grant Number 5 P03A 041 21.     

Linear versus logistic regression when the dependent variable is a dichotomy

The article argues against the popular belief that linear regression should not be used when the dependent variable is a dichotomy. The relevance of the statistical arguments against linear analyses, that the tests of significance are inappropriate and that one risk getting meaningless results, are disputed. Violating the homoscedasticity assumption seems to be of little practical importance, as an empirical comparison of results shows nearly identical outcomes for the two kinds of significance tests. When linear analysis of dichotomous dependent variables is seen as acceptable, there in many situations exist compelling arguments of a substantive nature for preferring this approach to logistic regression. Of special importance is the intuitive meaningfulness of the linear measures as differences in probabilities, and their applicability in causal (path) analysis, in contrast to the logistic measures.     

Optimal designs under two related models for interference

We consider two related models for interference: one having different directional neighbour effects and one with the same neighbour effects. We show that optimal designs for one model can be obtained from optimal designs for the other model. Acknowledgement. We are grateful to H. Kushner for advance notice of his work. SP was supported by the Isfahan University of Technology, Iran.     

Optimal extended complete block designs for dependent observations

Optimal designs under general dependence structures are usually difficult to specify theoretically or find algorithmically. However, they can sometimes be found for a specific dependence structure and a particular parameter value. In this paper, a class of generalized binary block designs with t treatments and b blocks of size k>t is considered. Each block consists of h consecutive complete blocks and, at the end, an incomplete block of size k−ht (if k > ht). For a suitable number of blocks, a universally optimal design is found for a first-order stationary autoregressive process with positive correlations. Optimal generalized binary designs and balanced block designs are also considered. Some constructions for a universally optimal design are described. A negative dependence parameter, and some other dependence structures, are also considered.     

Combining information for prediction in linear regression

This article deals with the prediction problem in linear regression where the measurements are obtained using k different devices or collected from k different independent sources. For the case of k=2, a Graybill-Deal type combined estimtor for the regression parameters is shown to dominate the individual least squares estimators under the covariance criterion. Two predictors ŷ _c and ŷ _p are proposed. ŷ _c is based on a combined estimator of the regression coefficient vector, and ŷ _p is obtained by combining the individual predictors from different models. Prediction mean square errors of both predictors are derived. It is shown that the predictor ŷ _p is better than the individual predictors for k≥2 and the predictor ŷ _c is better than the individual predictors for k=2. Numerical comparison between ŷ _c and ŷ _p shows that the former is superior to the latter for the case k=2.     

Optimal designs for generalized linear models with biased response

This paper is concerned with the search for locally optimal designs when the observations of the response variable arise from a weighted distribution in the exponential family. Locally optimal designs are derived for regression models in which the response follows a weighted version of Normal, Gamma, Inverse Gaussian, Poisson or Binomial distributions. Some conditions are given under which the optimal designs for the weighted and original (non-weighted) distributions are the same. An efficiency study is performed to find out the behavior of the D-optimal designs for the original distribution when they are used to estimate models with weighted distributions.     

Optimal fiscal zoning when the local government is a discriminating monopolist

Monopoly zoning by local governments has been studied extensively since the original article by White [1975, in: E. Mills and W. Oates, eds., Fiscal zoning and land use controls (Lexington Books, Lexington, MA)]. However, little or no attention has been paid to the possibility of monopolistic discrimination under such a regime. The question is: When will zoning multiple lot sizes be feasible given that buyers with different valuations of land cannot be coerced into purchasing particular lots? The literature on imperfect price discrimination by monopolists is brought to bear on this question. It is shown that the resulting zoning strategy may or may not involve multiple lot sizes. The factors determining the optimal strategy are studied in detail, with particular emphasis on the nature of the market for undeveloped land in the community.     

Optimal criteria and equivalence for nonregular fractional factorial designs

By considering the so-called algebraic orthogonality among experimental runs, this paper introduces a new criterion, minimum inner-product moment (MIPM), for general asymmetrical designs, and shows that MIPM is equivalent to the minimum moment aberration (MMA) criterion for the natural weights. Furthermore, the relationship between the generalized minimum aberration (GMA) and some model-dependent efficiency criteria is investigated by using the complex contrasts. Thus, two new justifications of GMA criterion is given from the points of view of orthogonality among experimental runs and design efficiency. They are generalizations of the related results of Butler (2003) and Cheng, Deng, and Tang (2002) for two-level factorial designs.     

Optimal designs for models with block-block resp. treatment-treatment correlations

Summary In this paper we study optimal designs assuming two special covariance structures of the observations, namely that the covariance between the observations depends only on the blocks resp. the treatments. We show that the weighted least squares estimator equals the ordinary least squares estimator. Then we prove that block-block correlations resp. treatment-treatment correlations do not have any influence on the A- and MV-optimality resp. A-optimality. For the study of the MV-optimality in the case of treatment-treatment correlations we use the idea of invariance to find optimal C-matrices.     

Exact D-optimal designs for first-order trigonometric regression models on a partial circle

Recently, various approximate design problems for low-degree trigonometric regression models on a partial circle have been solved. In this paper we consider approximate and exact optimal design problems for first-order trigonometric regression models without intercept on a partial circle. We investigate the intricate geometry of the non-convex exact trigonometric moment set and provide characterizations of its boundary. Building on these results we obtain a solution of the exact $D$ -optimal design problem. It is shown that the structure of the optimal designs depends on both the length of the design interval and the number of observations.     

Locally D- and c-optimal designs for Poisson and negative binomial regression models

The research on optimal experimental designs for nonlinear regression models is of great interest because these models are used to characterize chemical, biological or agricultural phenomena. Much of them involve an exponential decay. In this paper, locally D- and c-optimal designs are derived analytically for Poisson and negative binomial regression models.     

Optimal inference for instrumental variables regression with non-Gaussian errors

This paper is concerned with inference on the coefficient on the endogenous regressor in a linear instrumental variables model with a single endogenous regressor, nonrandom exogenous regressors and instruments, and i.i.d. errors whose distribution is unknown. It is shown that under mild smoothness conditions on the error distribution it is possible to develop tests which are “nearly” efficient in the sense of Andrews et al. (2006) when identification is weak and consistent and asymptotically optimal when identification is strong. In addition, an estimator is presented which can be used in the usual way to construct valid (indeed, optimal) confidence intervals when identification is strong. The estimator is of the two stage least squares variety and is asymptotically efficient under strong identification whether or not the errors are normal.