Path: news.mathworks.com!not-for-mail
From: "Bruno Luong" <b.luong@fogale.fr>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Is this kind of regression possible?
Date: Thu, 3 Jan 2008 07:45:26 +0000 (UTC)
Organization: FOGALE nanotech
Lines: 20
Message-ID: <fli3ql$h0r$1@fred.mathworks.com>
References: <fio8nm$6k4$1@fred.mathworks.com> <fior8p$asa$1@fred.mathworks.com> <flfcpq$nf8$1@fred.mathworks.com> <flh50h$kun$1@fred.mathworks.com> <flh7dg$8dc$1@fred.mathworks.com> <flh8oo$s3j$1@fred.mathworks.com> <flhb28$1n7$1@fred.mathworks.com>
Reply-To: "Bruno Luong" <b.luong@fogale.fr>
NNTP-Posting-Host: webapp-05-blr.mathworks.com
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Trace: fred.mathworks.com 1199346326 17435 172.30.248.35 (3 Jan 2008 07:45:26 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Thu, 3 Jan 2008 07:45:26 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 390839
Xref: news.mathworks.com comp.soft-sys.matlab:444145


"vicky " <vivek_mutalik@yahoo.com> wrote in message
<flhb28$1n7$1@fred.mathworks.com>...
> 
> Ive only Y (activity) and W's. If i give guessed values to
> all W's then i can use 'lsqnonlin' to obtain new coefficient
> values (W's) with respect to their activity Y for nonlinear
> least-squares problem. I was just thinking is there a method
> to do same for linear least-squares problem.

'lsqnonlin' is an *iterative solver* can be used for
non-linear AND linear least-square problem. 

The algebric formula I gave is a *direct method* for linear
least-square.

Both are able to take into account the first guess.

What is missing for you?

Bruno