MathGroup Archive 2006

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: multiple regression

  • To: mathgroup at smc.vnet.net
  • Subject: [mg68279] Re: multiple regression
  • From: Jean-Marc Gulliet <jeanmarc.gulliet at gmail.com>
  • Date: Mon, 31 Jul 2006 03:45:48 -0400 (EDT)
  • Organization: The Open University, Milton Keynes, UK
  • References: <eahsar$ooa$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

raj wrote:
> hai
> 
> i have exprimental datas, which consist of two inputs (say x1&x2) and
> corresponding ouput (say y). i have to relate these inputs and output
> to form quaderatic regressive equation as like Y=Ã?0+ Ã?1x1- Ã?2x1^2+
> Ã?3x2- Ã?4x2^2. where  Ã? is regressive co-efficient, which is also to
> be find. how i can get this type of equation using avilable data.
> kindly give the suitable technique.
> 
> thanks in advance
> n,kanagaraj
> 

The function *Fit* should do what you are looking for. See

http://documents.wolfram.com/mathematica/book/section-3.8.2

http://documents.wolfram.com/mathematica/functions/Fit

HTH,
Jean-Marc


  • Prev by Date: Re: Replacing Numbers in a List
  • Next by Date: Is there a good way to output parts of a Java object in Mathematica but STILL RETAIN THE OTHER INFO IN THE JAVA OBJECT?
  • Previous by thread: multiple regression
  • Next by thread: problem with Quaternion polynomial root solver