PlanetMath (more info)
 Math for the people, by the people. Sponsor PlanetMath
Encyclopedia | Requests | Forums | Docs | Wiki | Random | RSS  
Login
create new user
name:
pass:
forget your password?
Main Menu
Owner confidence rating: Very high Entry average rating: No information on entry rating
[parent] properties of spanning sets (Result)

Let $V$ be a vector space over a field $k$ . Let $S$ be a subset of $V$ . We denote $\Sp(S)$ the span of the set $S$ . Below are some basic properties of spanning sets.

  1. If $S\subseteq T$ , then $\Sp(S)\subseteq \Sp(T)$ . In particular, if $\Sp(S)=V$ , every superset of $S$ spans (generates) $V$ .
    Proof. If $v \in \Sp(S)$ , then $v=r_1v_1+\cdots +r_nv_n$ for $v_i\in S$ . But $v_i\in T$ by assumption. So $v\in \Sp(T)$ as well. If $\Sp(S)=V$ , and $S\subseteq T$ , then $V=\Sp(S)\subseteq \Sp(T) \subseteq V$ . $ \qedsymbol$
  2. If $S$ contains $0$ , then $\Sp(S-\lbrace 0\rbrace)=\Sp(S)$ .
    Proof. Let $T=S-\lbrace 0\rbrace$ . So $\Sp(T)\subseteq \Sp(S)$ by 1 above. If $v\in \Sp(S)$ , then $v=r_1v_1+\cdots + r_nv_n$ . If one of the $v_i$ 's, say $v_i$ , is $0$ , then $v=r_2v_2+\cdots +r_nv_n\in \Sp(T)$ . $ \qedsymbol$
  3. It is not true that if $S_1\supseteq S_2\supseteq \cdots$ is a chain of subsets, each spanning the same subspace $W$ of $V$ , so does their intersection.
    Proof. Take $V=\mathbb{R}^n$ , the Euclidean space in $n$ dimensions. For each $i=1,2,\ldots$ , let $S_i$ be the closed ball centered at the origin, with radius $1/i$ . Then $\Sp(S_i)=V$ . But the intersection of these $S_i$ 's is just the origin, whose span is itself, not $V$ . $ \qedsymbol$
  4. $S$ is a basis for $V$ iff $S$ is a minimal spanning set of $V$ . Here, minimal means that any deletion of an element of $S$ is no longer a spanning set of $V$ .
    Proof. If $S$ is a basis for $V$ , then $S$ spans $V$ and $S$ is linearly independent. Let $T$ be the set obtained from $S$ with $v\in S$ deleted. If $T$ spans $V$ , then $v$ can be written as a linear combination of elements in $T$ . But then $S=T\cup \lbrace v\rbrace$ would no longer be linearly independent, contradiction the assumption. Therefore, $S$ is minimal.

    Conversely, suppose $S$ is a minimal spanning set for $V$ . Furthermore, suppose that $S$ is linearly dependent. Let $0=r_1v_1+\cdots r_nv_n$ , with $r_1\ne 0$ . Then \begin{equation} v_1=s_2v_2+\cdots +s_nv_n, \end{equation}where $s_i=-r_i/r_1$ . So any linear combination of elements in $S$ involving $v_1$ can be replaced by a linear combination not involving $v_1$ through equation (1). Therefore $\Sp(S)=\Sp(S-\lbrace v\rbrace)$ . But this means that $S$ is not minimal, contrary to our assumption. Therefore, $S$ must be linearly independent. $ \qedsymbol$

Remark. All of the properties above can be generalized to modules over rings, except the last one, where the implication is only one-sided: basis implying minimal spanning set.




"properties of spanning sets" is owned by CWoo.
(view preamble | get metadata)

View style:


This object's parent.
Log in to rate this entry.
(view current ratings)

Cross-references: implication, rings, modules, properties, equation, linearly dependent, conversely, contradiction, linear combination, linearly independent, element, deletion, spanning set, minimal, iff, basis, radius, origin, closed ball, dimensions, Euclidean space, intersection, subspace, spanning, chain, contains, generates, superset, span, subset, field, vector space

This is version 4 of properties of spanning sets, born on 2008-05-28, modified 2008-05-29.
Object id is 10634, canonical name is PropertiesOfSpanningSets.
Accessed 598 times total.

Classification:
AMS MSC15A03 (Linear and multilinear algebra; matrix theory :: Vector spaces, linear dependence, rank)
 16D10 (Associative rings and algebras :: Modules, bimodules and ideals :: General module theory)

Pending Errata and Addenda
None.
Discussion
Style: Expand: Order:
forum policy

No messages.

Interact
post | correct | update request | add example | add (any)