Multilevel Augmented Pairs Second-Order Response Surface Designs and Their Robustness to Missing Data

Missing observations can occur even in a well-planned experiment. The effect of missing observations can be much more serious when the design is saturated or near saturated. The levels of factor settings that make a design more robust to missing observations are of great importance in the sense that the loss for missing observations becomes minimum. In this study, new augmented pairs minimax loss designs are constructed, which are more robust to one missing design point than the augmented pairs designs presented by Morris (2000 Morris , M. D. ( 2000 ). A class of three-level experimental designs for response surface modeling . Technometrics 42 : 111 – 121 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]). New designs are compared with augmented pairs designs, central composite designs, and small composite designs under generalized scaled standard deviations. The model used is also studied for the regression coefficient estimates.     

Robust Bayesian methodology with applications in credibility premium derivation and future claim size prediction

Robust Bayesian methodology deals with the problem of explaining uncertainty of the inputs (the prior, the model, and the loss function) and provides a breakthrough way to take into account the input’s variation. If the uncertainty is in terms of the prior knowledge, robust Bayesian analysis provides a way to consider the prior knowledge in terms of a class of priors \(\varGamma \) and derive some optimal rules. In this paper, we motivate utilizing robust Bayes methodology under the asymmetric general entropy loss function in insurance and pursue two main goals, namely (i) computing premiums and (ii) predicting a future claim size. To achieve the goals, we choose some classes of priors and deal with (i) Bayes and posterior regret gamma minimax premium computation, (ii) Bayes and posterior regret gamma minimax prediction of a future claim size under the general entropy loss. We also perform a prequential analysis and compare the performance of posterior regret gamma minimax predictors against the Bayes predictors.     

Bayes minimax ridge regression estimators

The problem of estimating of the vector β of the linear regression model y = Aβ + ? with ? ～ N_p(0, σ²I_p) under quadratic loss function is considered when common variance σ² is unknown. We first find a class of minimax estimators for this problem which extends a class given by Maruyama and Strawderman (2005 Maruyama, Y., and W. E. Strawderman. 2005. A new class of generalized Bayes minimax ridge regression estimators. Annals of Statistics 33:1753–70.[Crossref], [Web of Science ®] , [Google Scholar]) and using these estimators, we obtain a large class of (proper and generalized) Bayes minimax estimators and show that the result of Maruyama and Strawderman (2005 Maruyama, Y., and W. E. Strawderman. 2005. A new class of generalized Bayes minimax ridge regression estimators. Annals of Statistics 33:1753–70.[Crossref], [Web of Science ®] , [Google Scholar]) is a special case of our result. We also show that under certain conditions, these generalized Bayes minimax estimators have greater numerical stability (i.e., smaller condition number) than the least-squares estimator.     

D-optimal Factorial Designs under Generalized Linear Models

Generalized linear models (GLMs) have been used widely for modeling the mean response both for discrete and continuous random variables with an emphasis on categorical response. Recently Yang, Mandal and Majumdar (2013 Yang, J., Mandal, A., Majumdar, D. (2013). Optimal designs for 2^k factorial experiments with binary response. Technical Report, Available at: http://arxiv.org/pdf/1109.5320v4.pdf. [Google Scholar]) considered full factorial and fractional factorial locally D-optimal designs for binary response and two-level experimental factors. In this article, we extend their results to a general setup with response belonging to a single-parameter exponential family and for multilevel predictors.     

Minimax sequential tests of composite hypotheses on the drift of a Wiener process

A Wiener process with unknown drift parameter μ is, beginning at O, observed continuously and one has to decide between the hypotheses μ≤0 and μ>0. For loss functions of the form sμ^r and linear cost functions one wants to determine a minimax sequential test. Generalizing the results of DeGroot (1960) a minimax test in the class of all symmetrical SPRT’s is given in an explicit form. On the other hand it is shown that this SPRT is, in general, no longer minimax in the class of all sequential tests.     

On minimaxity of the normal precision matrix estimator of Krishnamoorthy and Gupta

We consider the orthogonally invariant estimation problem of the inverse of the scale matrix of Wishart distribution using Stein's loss (entropy loss). In this problem Krishnamoorthy and Gupta [2] Krishnamoorthy, K. and Gupta, A. K. (1989). Improved minimax estimation of a normal precision matrix. Canad. J. Statist., 17: 91–102. [Crossref], [Web of Science ®] , [Google Scholar] proposed an estimator and showed its good performance in a Monte Carlo simulation. They conjectured their estimator is minimax. Perron [3] Perron, F. (1997). On a conjecture of Krishnamoorthy and Gupta. J. Multivariate Anal., 62: 110–120.  [Google Scholar] proved its minimaxity for p?=?2. In this paper we prove it for p?=?3 by using a new method.     

Minimax estimators for the lower-bounded scale parameter of a location-scale family of distributions

This article is concerned with the minimax estimation of a scale parameter under the quadratic loss function where the family of densities is location-scale type. We obtain results for the case when the scale parameter is bounded below by a known constant. Implications for the estimation of a lower-bounded scale parameter of an exponential distribution are presented under unknown location. Furthermore, classes of improved minimax estimators are derived for the restricted parameter using the Integral Expression for Risk Difference (IERD) approach of Kubokawa (1994 Kubokawa, T. (1994). A unified approach to improving equivariant estimators. Ann. Stat. 22:290–299.[Crossref], [Web of Science ®] , [Google Scholar]). These classes are shown to include some existing estimators from literature.     

Extensions of D-optimal minimal designs for symmetric mixture models

The purpose of mixture experiments is to explore the optimum blends of mixture components, which will provide the desirable response characteristics in finished products. D-optimal minimal designs have been considered for a variety of mixture models, including Scheffé's linear, quadratic, and cubic models. Usually, these D-optimal designs are minimally supported since they have just as many design points as the number of parameters. Thus, they lack the degrees of freedom to perform the lack-of-fit (LOF) tests. Also, the majority of the design points in D-optimal minimal designs are on the boundary: vertices, edges, or faces of the design simplex. In this article, extensions of the D-optimal minimal designs are developed for a general mixture model to allow additional interior points in the design space to enable prediction of the entire response surface. Also a new strategy for adding multiple interior points for symmetric mixture models is proposed. We compare the proposed designs with Cornell (1986 Cornell, J.A. (1986). A comparison between two ten-point designs for studying three-component mixture systems. J. Qual. Technol. 18(1):1–15.[Taylor &; Francis Online], [Web of Science ®] , [Google Scholar]) two 10-point designs for the LOF test by simulations.     

Efficient Mixture Design Fitting Quadratic Surface with Quantile Responses Using First-degree Polynomial

This study considers efficient mixture designs for the approximation of the response surface of a quantile regression model, which is a second degree polynomial, by a first degree polynomial in the proportions of q components. Instead of least squares estimation in the traditional regression analysis, the objective function in quantile regression models is a weighted sum of absolute deviations and the least absolute deviations (LAD) estimation technique should be used (Bassett and Koenker, 1982 Bassett, G., Koenker, R. (1982). An empirical quantile function for linear models with i.i.d. errors. Journal of the American Statistical Association 77:407–415.[Taylor &; Francis Online], [Web of Science ®] , [Google Scholar]; Koenker and Bassett, 1978 Koenker, R., Bassett, G. (1978). Regression quantiles. Econometrica 46(1):33–50.[Crossref], [Web of Science ®] , [Google Scholar]). Therefore, the standard optimal mixture designs like the D-optimal or A-optimal mixture designs for the least squared estimation are not appropriate. This study explores mixture designs that minimize the bias between the approximated 1st-degree polynomial and a 2nd-degree polynomial response surfaces by the LAD estimation. In contrast to the standard optimal mixture designs for the least squared estimation, the efficient designs might contain elementary centroid design points of degrees higher than two. An example of a portfolio with five assets is given to illustrate the proposed efficient mixture designs in determining the marginal contribution of risks by individual assets in the portfolio.     

Conditions of Slope-Rotatability for Third-Order Polynomial Regression Models

In experimental design for response surface analysis, it is sometimes of interest to estimate the difference of responses at two points. If differences at points close together are involved, the design that reliably estimates the slope of the response surface is important. In particular, Hader and Park (1978 Hader , R. J. , Park , S. H. ( 1978 ). Slope-rotatable central composite designs . Technometrics 20 : 413 – 417 .[Taylor & Francis Online], [Web of Science ®] , [Google Scholar]) suggested the concept of slope-rotatability and studied slope rotatable central composite designs. Until now, many response surface designs including central composite designs have been suggested for fitting second order response surface models. However, we often need to fit third-order polynomial regression models. In this article, we suggest extended central composite designs (ECCDs) to fit third-order models and find the necessary and sufficient conditions for slope-rotatability over all directions in the third-order polynomial models.     

A Note on the Minimax Solution for the Two-Stage Group Testing Problem

Group testing is an active area of current research and has important applications in medicine, biotechnology, genetics, and product testing. There have been recent advances in design and estimation, but the simple Dorfman procedure introduced by R. Dorfman in 1943 Dorfman, R. (1943), The Detection of Defective Members of Large Populations, The Annals of Mathematical Statistics, 14, 436–440.[Crossref] , [Google Scholar] is widely used in practice. In many practical situations, the exact value of the probability p of being affected is unknown. We present both minimax and Bayesian solutions for the group size problem when p is unknown. For unbounded p, we show that the minimax solution for group size is 8, while using a Bayesian strategy with Jeffreys’ prior results in a group size of 13. We also present solutions when p is bounded from above. For the practitioner, we propose strong justification for using a group size of between 8 and 13 when a constraint on p is not incorporated and provide useable code for computing the minimax group size under a constrained p.     

Some New Series of Partially Balanced Residual Treatment Effects Designs

Blaisdell and Raghavarao (1980 Blaisdell , E. A. , Raghavarao , D. ( 1980 ). Partially balanced change-over designs based on m-associate class PBIB designs . JRSS B 42 : 334 – 338 . [Google Scholar] 1985 Blaisdell , E. A. , Raghavarao , D. ( 1985 ). Efficiency bounds for partially balanced change-over designs based on m-associate class PBIB designs . JRSS B 47 : 132 – 135 . [Google Scholar]) introduced partially balanced residual treatment effects designs and gave methods of construction of these designs. They also gave efficiencies for the estimated direct and residual elementary treatment contrasts of these designs. In this paper, we construct some new series of partially balanced residual treatment effects designs and also give their efficiencies for the estimated direct and residual elementary treatment contrasts.     

Rotatability is a Sufficient Condition for A- and D-Rotatability

Designs for estimating axial slopes of a response surface are considered. A design is said to be A- (D-) rotatable if the trace (determinant) of variance–covariance matrix of the estimated slopes at a point in the factor space is a function of the distance of the point from the center of the design. It is shown that for any order d and any dimension k, a design which is rotatable in the Box and Hunter (1957 Box , G. E. P. , Hunter , J. S. ( 1957 ). Multifactor experimental designs for exploring response surfaces . Annals of Mathematical Statistics 28 : 195 – 241 .[Crossref] , [Google Scholar]) sense is automatically A- and D-rotatable. Some specific examples illustrating the validity of this general result are provided.     

On Some Optimal Bayesian Nonparametric Rules for Estimating Distribution Functions

In this paper, we present a novel approach to estimating distribution functions, which combines ideas from Bayesian nonparametric inference, decision theory and robustness. Given a sample from a Dirichlet process on the space (𝒳, A), with parameter η in a class of measures, the sampling distribution function is estimated according to some optimality criteria (mainly minimax and regret), when a quadratic loss function is assumed. Estimates are then compared in two examples: one with simulated data and one with gas escapes data in a city network.     

Analyzing Supersaturated Designs by Means of an Information Based Criterion

The cost and time consumption of many industrial experimentations can be reduced using the class of supersaturated designs since this can be used for screening out the important factors from a large set of potentially active variables. A supersaturated design is a design for which there are fewer runs than effects to be estimated. Although there exists a wide study of construction methods for supersaturated designs, their analysis methods are yet in an early research stage. In this article, we propose a method for analyzing data using a correlation-based measure, named as symmetrical uncertainty. This method combines measures from the information theory field and is used as the main idea of variable selection algorithms developed in data mining. In this work, the symmetrical uncertainty is used from another viewpoint in order to determine more directly the important factors. The specific method enables us to use supersaturated designs for analyzing data of generalized linear models for a Bernoulli response. We evaluate our method by using some of the existing supersaturated designs, obtained according to methods proposed by Tang and Wu (1997 Tang , B. , Wu , C. F. J. (1997). A method for constructing supersaturated designs and its E(s ²)-optimality. Canadian Journal of Statistics 25:191–201.[Crossref], [Web of Science ®] , [Google Scholar]) as well as by Koukouvinos et al. (2008 Koukouvinos , C. , Mylona , K. , Simos , D. E. ( 2008 ). E(s ²)-optimal and minimax-optimal cyclic supersaturated designs via multi-objective simulated annealing . Journal of Statistical Planning and Inference 138 : 1639 – 1646 .[Crossref], [Web of Science ®] , [Google Scholar]). The comparison is performed by some simulating experiments and the Type I and Type II error rates are calculated. Additionally, Receiver Operating Characteristics (ROC) curves methodology is applied as an additional statistical tool for performance evaluation.     

Optimum Mixture Design Using Deficiency Criterion

In a mixture experiment the measured response is assumed to depend only on the relative proportion of ingredients or components present in the mixture. Scheffe (1958 Scheffe , H. ( 1958 ). Experiments with mixtures . J. R. Statist. Soc. B 20 : 344 – 360 . [Google Scholar], 1963 Scheffe , H. ( 1963 ). Simplex – centroid design for experiments with mixtures . J. R. Statist. Soc. B 25 : 235 – 263 . [Google Scholar]) first systematically considered this problem and introduced different models and designs suitable in such situations. Optimum designs for the estimation of parameters of different mixture models are available in the literature. The problem of estimating the optimum proportion of mixture components is of great practical importance. Pal and Mandal (2006 Pal , M. , Mandal , N. K. ( 2006 ). Optimum designs for optimum mixtures . Statist. Probab. Lett. 76 ( 13 ): 1369 – 1379 .[Crossref], [Web of Science ®] , [Google Scholar], 2007 Pal , M. , Mandal , N. K. (2007). Optimum mixture design via equivalence theory. communicated. JCISS 32:107–126. [Google Scholar]) attempted to find a solution to this problem by adopting a pseudo-Bayesian approach and using the trace criterion. Subsequently, Pal and Mandal (2008 Pal , M. , Mandal , N. K. ( 2008 ). Minimax designs for optimum mixtures . To appear in Statist. Probab. Lett.  [Google Scholar]) solved the problem using minimax criterion. In this article, the deficiency criterion due to Chatterjee and Mandal (1981 Chatterjee , S. K. , Mandal , N. K. ( 1981 ). Response surface designs for estimating the optimal point . Calcutta Statist. Assoc. Bull. . 30 : 145 – 169 .[Crossref] , [Google Scholar]) has been used as a measure for comparing the performance of competing designs.