The Concept of Majorization in Experimental Designs
نویسنده
چکیده
The information of interest is contained in the variance matrix, σ − = 2 1 V Q , or equivalently in the information matrix Q . The problem of finding estimators minimizing a decreasing convex functional of the variance matrix or maximizing an increasing concave functional of the information matrix is crucial in statistics. There are several optimality criteria in bibliography. We explain how the concept of Majorization can be used to investigate optimal experimental designs.
منابع مشابه
Latin-majorization and its linear preservers
In this paper we study the concept of Latin-majorizati-\on. Geometrically this concept is different from other kinds of majorization in some aspects. Since the set of all $x$s Latin-majorized by a fixed $y$ is not convex, but, consists of union of finitely many convex sets. Next, we hint to linear preservers of Latin-majorization on $ mathbb{R}^{n}$ and ${M_{n,m}}$.
متن کاملLinear preservers of Miranda-Thompson majorization on MM;N
Miranda-Thompson majorization is a group-induced cone ordering on $mathbb{R}^{n}$ induced by the group of generalized permutation with determinants equal to 1. In this paper, we generalize Miranda-Thompson majorization on the matrices. For $X$, $Yin M_{m,n}$, $X$ is said to be Miranda-Thompson majorized by $Y$ (denoted by $Xprec_{mt}Y$) if there exists some $Din rm{Conv(G)}$ s...
متن کاملMajorization framework for balancedlattice designs
This paper aims to generalize and unify classical criteria for comparisons of balanced lattice designs, including fractional factorial designs, supersaturated designs and uniform designs. We present a general majorization framework for assessing designs, which includes a stringent criterion of majorization via pairwise coincidences and flexible surrogates via convex functions. Classical orthogo...
متن کاملMatrix majorization Geir Dahl
We study the conceptmatrix majorization: for two real matrices A and B having m rows we say that A majorizes B if there is a row-stochastic matrix X with AX = B. A special case is classical notion of vector majorization. Several properties and characterizations of matrix majorization are given. Moreover, interpretations of the concept in mathematical statistics are discussed and some combinator...
متن کاملWeak matrix majorization
Given X, Y ∈ Rn×m we introduce the following notion of matrix majorization, called weak matrix majorization, X w Y if there exists a row-stochastic matrix A ∈ Rn×n such that AX = Y, and consider the relations between this concept, strong majorization ( s ) and directional majorization ( ). It is verified that s⇒ ⇒ w , but none of the reciprocal implications is true. Nevertheless, we study the i...
متن کامل