HOCcholesky

# HOCcholesky - Solving Systems of Linear Equations by...

This preview shows pages 1–2. Sign up to view the full content.

Page 1 of 5 5/14/2003 cholesky.doc Last Revised: November 5, 2002, by R. F. Taylor. Ref: Numerical Analysis: Mathematics of Scientific computing , Third Edition, D. Kincaid and W. Cheney, Brooks/Cole, 2002, ISBN 0-534- 38905-8, page 158. 1.0 When to Apply the Cholesky Method The Cholesky method is an effective alternative to LU decomposition when the n by n A matrix is real, symmetric, and positive definite. It is easy to determine when a matrix is symmetric since A T = A in this case. A matrix is positive definite if x T Ax > 0 for all choices of real n- vectors x . Note that x T Ax is just a single positive number which depends on x and A . This is not a very useful definition to test for positive definiteness since a positive number must result for all x vectors. A theorem in matrix algebra states that a matrix is positive definite if the determinants of the upper left principal submatrices are all positive. For example, consider the following matrix (Matlab notation): A=[4– 2– 4 ; -2 10 14; -4 14 56] To use the above theorem, we need to show that the following submatrices all have positive determinants: A 1=[4] A 2=[4– 2 ; -2 10] A 3=A=[4– 4 ; -2 10 14; -4 14 56] In this case we have det(A1)=4>0 , det(A2) = 16 > 0, and det(A3) = det(A) = 1296 and hence the matrix is positive definite.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 04/26/2010 for the course CEG 616 taught by Professor Taylor during the Winter '10 term at Wright State.

### Page1 / 5

HOCcholesky - Solving Systems of Linear Equations by...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online