2009
DOI: 10.13001/1081-3810.1302
|View full text |Cite
|
Sign up to set email alerts
|

Generating potentially nilpotent full sign patterns

Abstract: Abstract.A sign pattern is a matrix with entries in {+, −, 0}. A full sign pattern has no zero entries. The refined inertia of a matrix pattern is defined and techniques are developed for constructing potentially nilpotent full sign patterns. Such patterns are spectrally arbitrary. These techniques can also be used to construct potentially nilpotent sign patterns that are not full, as well as potentially stable sign patterns.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

1
19
0

Year Published

2010
2010
2016
2016

Publication Types

Select...
8

Relationship

2
6

Authors

Journals

citations
Cited by 21 publications
(20 citation statements)
references
References 17 publications
1
19
0
Order By: Relevance
“…In [9] the question was raised as to whether or not a sign pattern that is an rIAP must be a SAP. In Section 2 we address this question for irreducible zero-nonzero patterns and exhibit such a pattern of smallest order that is an rIAP but not a SAP.…”
Section: Lemma 12 ([8 Proposition 2])mentioning
confidence: 99%
“…In [9] the question was raised as to whether or not a sign pattern that is an rIAP must be a SAP. In Section 2 we address this question for irreducible zero-nonzero patterns and exhibit such a pattern of smallest order that is an rIAP but not a SAP.…”
Section: Lemma 12 ([8 Proposition 2])mentioning
confidence: 99%
“…Similarly, a pattern A is said to be PPI + if A is potentially purely imaginary and G(A) allows a positive loop, allows a negative loop, and allows a negative 2-cycle. The property PN + was defined in [14] for sign patterns but explicitly required that G(A) has at least two loops. In our definition, it is possible that A can be PN + (or PPI + ) with the property that G(A) has exactly one loop with sign #.…”
Section: Cavers and S Fallatmentioning
confidence: 99%
“…In our definition, it is possible that A can be PN + (or PPI + ) with the property that G(A) has exactly one loop with sign #. The definition of PN + from [14] is motivated in part by Theorem 2.4 below, which is a generalization of [5,Lemma 5.1]. The definition of PN + in this paper is motivated by Corollary 2.8 and is equivalent to the definition in [14] for {0, +, −, * }-patterns.…”
Section: Cavers and S Fallatmentioning
confidence: 99%
See 1 more Smart Citation
“…As defined in [7], the refined inertia ri(A) of a real n × n matrix A is the ordered 4-tuple (n + , n − , n z , 2n p ) such that n + (resp., n − ) is the number of eigenvalues (including multiplicities) of A with positive (resp., negative) real part, and n z (resp., 2n p ) is the number of zero eigenvalues (resp., nonzero pure imaginary eigenvalues) of A. Here n + + n − + n z + 2n p = n. The inertia of A is (n + , n − , n z + 2n p ), thus the refined inertia subdivides those eigenvalues with zero real part and distinguishes between those that are exactly zero and those that are nonzero.…”
mentioning
confidence: 99%