Mathematica 9 is now available
Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2007
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2007

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

Search the Archive

Re: HELP IN RECURSIVE ESTIMATION WITH MATHEMATICA

  • To: mathgroup at smc.vnet.net
  • Subject: [mg78122] Re: HELP IN RECURSIVE ESTIMATION WITH MATHEMATICA
  • From: David Bailey <dave at Remove_Thisdbailey.co.uk>
  • Date: Sat, 23 Jun 2007 07:12:48 -0400 (EDT)
  • References: <f5gall$gbj$1@smc.vnet.net>

Cecilia Maria Caruso wrote:
> Hallo, 
> I'm Cecilia Caruso, an italian management engineer who is going to take her II level degree. I'm studying for my final work in Statistical Quality Control and i need some informations.
> 
> My teacher used a sigmoidal model to study a process and she used fitting tools of Mathematica to manage and study the data. 
> Now i have to perform CUSUM and CUSUMQ tests on it, to test the constancy of parameters vector along the time. 
> 
> I must use a "recursive least squares" estimation method to perform the tests. I must use this method because the tests involve recursive residuals. I have a "Mathematica 5.1" version and i can't see if it's possible to perform these tests with your software. 
> Can you help me? Is it possible to perform these tests with Mathematica 5.1?
> Thank you so much
> Greetings,
> Cecilia Maria Caruso
> 
> 
> ------------------------------------------------------
> Scegli infostrada: ADSL gratis per tutta l=92estate e telefoni senza canone Telecom
> http://click.libero.it/infostrada
> 
> 
> 
This is a bit like looking at a hammer and asking if it would be 
suitable for constructing a particular kind of garden shed!

It is possible that someone has a package that does what you require, 
but otherwise you will have to program Mathematica to do what you want.

I would be very cautious about using any computerised mathematical 
method without having a reasonable understanding of what it is trying to 
do, and how it should be applied to your application. Don't just use 
some software like a black box!

David Bailey
http://www.dbaileyconsultancy.co.uk


  • Prev by Date: Re: Normally Distributed Random Matrix
  • Next by Date: Re: Combination List
  • Previous by thread: HELP IN RECURSIVE ESTIMATION WITH MATHEMATICA
  • Next by thread: Graphics card upgrade