Permutation matrix and PA = LDU

In summary, the conversation discusses finding the PA = LDU factorizations for a given matrix A and the use of permutation matrices in the process. It is mentioned that using different permutation matrices can result in different LDU factorizations due to the fact that the input to the algorithm is different. The use of a permutation matrix P is also clarified, as it is usually determined by the LU decomposition algorithm with pivoting, not chosen beforehand.
  • #1
Dafe
145
0

Homework Statement


Find the PA = LDU factorizations for:



A = [tex]\left[ \begin{array}{ccc} 0 & 1 & 1 \\ 1 & 0 & 1 \\ 2 & 3 & 4 \end{array} \right] [/tex]

The author chooses a permutation matrix :

P = [tex]\left[ \begin{array}{ccc} 0 & 1 & 0 \\ 1 & 0 & 0 \\ 0 & 0 & 1 \end{array} \right] [/tex]

If I do the same I obviously get the same factorization as in the book.

If I choose a permutation matrix,

P = [tex]\left[ \begin{array}{ccc} 0 & 0 & 1 \\ 1 & 0 & 0 \\ 0 & 1 & 0 \end{array} \right] [/tex]

I do not get the same factorization. Should I get the same answers even though I'm arranging the equations somewhat different, or am I missing something here?

Help is greatly appreciated, thanks!
 
Physics news on Phys.org
  • #2
If LU decomposition is performed with pivoting, P is the outcome of the decomposition process depending on pivoting. So, it seems odd that you or the author would be choosing P up front.

Maybe you are trying to study the affect on LU w/o pivoting by swapping rows of the original matrix? In any event, I would think using different P's would result in different LU factorizations because the matrix to be factored changes. I don't think two different matrices can have the same LU decomposition.
 
  • #3
Hi,

Say that I want to do elimination on the above matrix A.
As one possibility, I could start by exchanging row 1 with row 3 such that the integer 2 is in the first pivot position.

I would then perform elimination until I have a upper triangular matrix U.
I get the matrix U by multiplying A with a permutation matrix P, and two elementary matrices, say E and F.

[tex] U = FEPA [/tex]

Then L would be the product of the inverse of E and F.

[tex]L= E^{-1}F^{-1}[/tex]

I could also have started the elimination process by exchanging row 1 with row 2, which would have yielded a different permutation matrix, and also a different factorization LDU.

Am I totally off here?

Thank you for your time!
 
  • #4
Correct, you'd have two different LDU factorizations because you are factoring two different matrices. Even though you are starting from the same matrix in your two scenarios, the input to the LDU factorization algorithm is a different permutation of the original matrix in each case.

What makes this a bit confusing is your use of the permutation matrix P. Like I said, P is usually determined by the LU decomposition algorithm with pivoting, not something you choose ahead of time. Though perfectly valid, what you are doing is permuting the matrix with a P of your choice before performing LU decomposition w/o pivoting.
 
Last edited:
  • #5
I understand, thank you :)
 

1. What is a permutation matrix?

A permutation matrix is a square matrix with exactly one entry of 1 in each row and column, and all other entries are 0. It is used to represent a permutation of the rows or columns of another matrix.

2. How is a permutation matrix used in PA = LDU decomposition?

In PA = LDU decomposition, a permutation matrix is used to reorder the rows of a matrix A to create a lower triangular matrix L and an upper triangular matrix U. This allows for easier calculation of the diagonal matrix D, resulting in a decomposition of A into PA = LDU form.

3. What is the significance of PA = LDU decomposition?

PA = LDU decomposition is a useful tool in linear algebra as it allows for the simplification and solving of systems of linear equations. It also provides insights into the properties of a matrix, such as its rank and determinant.

4. Can a permutation matrix be used for any matrix?

Yes, a permutation matrix can be used for any square matrix. However, it is most commonly used in the context of PA = LDU decomposition for square matrices.

5. How is PA = LDU decomposition related to LU decomposition?

PA = LDU decomposition is an extension of LU decomposition, where the permutation matrix P is added to the equation. This allows for more flexibility and ease in solving systems of linear equations, as the permutation matrix can help avoid division by 0 and other issues that may arise in the calculation of L and U.

Similar threads

  • Calculus and Beyond Homework Help
Replies
2
Views
88
  • Calculus and Beyond Homework Help
Replies
6
Views
530
  • Calculus and Beyond Homework Help
Replies
1
Views
1K
  • Calculus and Beyond Homework Help
Replies
0
Views
151
  • Calculus and Beyond Homework Help
Replies
6
Views
295
  • Calculus and Beyond Homework Help
Replies
9
Views
1K
  • Calculus and Beyond Homework Help
Replies
7
Views
3K
  • Calculus and Beyond Homework Help
Replies
7
Views
1K
  • Calculus and Beyond Homework Help
Replies
3
Views
328
  • Linear and Abstract Algebra
Replies
5
Views
941
Back
Top