Prove the Matrix Inequalities Theorem

Click For Summary

Homework Help Overview

The discussion revolves around proving the Matrix Inequalities Theorem, which involves relationships between matrices A, B, C, and D. The theorem states conditions under which one matrix is less than or equal to another, focusing on properties of matrix inequalities.

Discussion Character

  • Conceptual clarification, Assumption checking

Approaches and Questions Raised

  • The original poster attempts to prove the theorem using specific matrix entries and inequalities. Some participants question the definition of the inequality for matrices and whether the properties assumed for real numbers apply to matrices. Others suggest that understanding the definitions is crucial for the proof.

Discussion Status

Participants are exploring the definitions and properties of matrix inequalities. Some guidance has been offered regarding the need for clarity on the definitions and the implications of the inequalities. There is an ongoing examination of the assumptions made in the original attempt.

Contextual Notes

There is a noted lack of reference material for the original poster, as their textbook does not cover the properties of matrix inequalities in depth. This has led to uncertainty about the definitions and the validity of their approach.

csc2iffy
Messages
74
Reaction score
0

Homework Statement



Theorem:
Suppose A, B, C, and D are matrices of the same size. Then
a) If A ≤ B and B ≤ C, then A ≤ C
b) If A ≤ B and C ≤ D, then A + C ≤ B + D
c) If A ≤ B, then cA ≤ cB for any positive constant c and cA >= cB for any negative constant c

Prove this theorem. Must use arbitrary matrices, one where both the size and entries are specified as variables

Homework Equations





The Attempt at a Solution



Let A=[ai,j], B=[bi,j], C=[ci,j], D=[di,j]

a) Let (1) [ai,j] ≤ [bi,j]
(2) [bi,j] ≤ [ci,j]
Adding (1) and (2), we get
[ai,j] + [bi,j] ≤ [bi,j] + [ci,j]
Subtracting [bi,j] from both sides,
[ai,j] ≤ [ci,j]
Therefore A ≤ C

b) Let (1) [ai,j] ≤ [bi,j]
(2) [ci,j] ≤ [di,j]
Adding (1) and (2), we get
[ai,j] + [ci,j] ≤ [bi,j] + [di,j]
Therefore A + C ≤ B + D

c) Not really sure how to do c?
 
Last edited by a moderator:
Physics news on Phys.org
what is the definition of <= for matrices?
 
-----
 
Last edited:
-----
 
Last edited:
First, tell us what it means to say matrix A ≤ matrix B. You have to know your definitions in math.
 
OK. this proof is in my linear programming class. I cannot remember what this means.. she did not give us a recap on inequalities of matrices
 
Last edited by a moderator:
Ya, you are on the right track and "c" is related to inequalities in algebra.
 
csc2iffy said:
OK. this proof is in my linear programming class. I cannot remember what this means.. she did not give us a recap on inequalities of matrices
Then how can you possibly know what properties matrix inequalities have? Like how do you know that A+B ≤ B+C implies A ≤ C? It's true for real numbers, but you can't automatically assume it holds for matrices.

Don't you have a textbook you can consult for basic definitions?
 
  • #10
nope my textbook just goes into the matrices of LP problems, not their properties :(
i just assumed it worked with matrices as it does with regular numbers, since any entry in A or B is just a number... so my attempt is completely wrong?
 
Last edited by a moderator:
  • #11
Well, it depends. If I had to guess, perhaps A, B, C, D are square matrices of the same size and the ordering is on the determinants of those matrices?

Could you post more context?
 
  • #13
-----
 
Last edited:
  • #14
In (a), for all i,j, ai,j ≤ bi,j, and bi,j ≤ ci,j implies what about the relationship between the elements of A and C?

Proceed similarly for the other parts.

dirk_mec1's link may be helpful. micromass' link delves a bit too deep to be useful.

If you want, you can go deeper with respect to this problem. The definition is an example of a http://en.wikipedia.org/wiki/Partial_order
 
  • #15
-----
 
Last edited:
  • #16
Notation-wise, I'd remove the square brackets because you're talking about the specific elements of each matrix. Note that they don't use the square brackets in the definition you cited. Otherwise, they look good.
csc2iffy said:
Since A ≤ B, then ai,j ≤ bi,j for all i,j entries of A and B.

Since B ≤ C, then bi,j ≤ ci,j for all i,j entries of B and C.

For all i,j entries, since ai,j ≤ bi,j and bi,j ≤ ci,j, then, by the [STRIKE]transient[/STRIKE] transitive property of inequalities, ai,j ≤ ci,j.

Therefore A ≤ C.
 

Similar threads

Replies
20
Views
2K
Replies
1
Views
960
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
2K
  • · Replies 12 ·
Replies
12
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K