Similitudes and the \(\ell_ 1\)-norm (Q1923200)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Similitudes and the \(\ell_ 1\)-norm |
scientific article; zbMATH DE number 931931
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Similitudes and the \(\ell_ 1\)-norm |
scientific article; zbMATH DE number 931931 |
Statements
Similitudes and the \(\ell_ 1\)-norm (English)
0 references
25 November 1996
0 references
Let \(C= \{c\in \mathbb{R}^n: \sum^n_{i=1} c_i= 0\}\), and let \(\alpha>0\). The author characterizes the matrices \(Q:C \to C\) satisfying \(|cQ |= \alpha |c |\) for all \(c \in C\), where \(|\cdot |\) denotes the \(\ell_1\)-norm. They are precisely of the form \(Q= \pm \alpha P+F\), where \(P\) is a permutation matrix, and \(F\) has all rows identical. In terms of the coefficient of ergodicity \({\mathcal T} (A) = \max \{|cA |/ |c |: c\in C,\;c \neq 0\}\), the characterization is that \({\mathcal T} (AQ) = {\mathcal T} (QA) = \text{const.} {\mathcal T} (A)\) for all stochastic matrices \(A\).
0 references
coefficient of ergodicity
0 references
stochastic matrices
0 references