Re: Singular Value Decomposition
- To: mathgroup at smc.vnet.net
- Subject: [mg6707] Re: Singular Value Decomposition
- From: Julian Stoev <stoev at iname.com>
- Date: Fri, 11 Apr 1997 01:50:20 -0400 (EDT)
- Organization: Seoul National University, Republic of Korea
- Sender: owner-wri-mathgroup at wolfram.com
On 8 Apr 1997, DR JOHN C ERB wrote: |I have a problem in which I have an overdetermined set of linear |equations. According to the book Numerical Recipes (Chapter 2, |Section 6), I can use the Singular Value Decomposition (SVD) to find |a least-squares fit to find a solution. | |Does anyone know of a reference, WITH EXAMPLES, which will "lead me |by the hand" on how to do this? | |Thank you. |John C. Erb Just look in the same book for "Linear regression". There you will find a lot for Least Squares. LU decomposition is just a method for solving some class of matrix problems. It seems to me that you are not interested how exactly your system will be solved. If this does not help, it's better to go in some university for a while. ============ Julian Stoev