Solved

multivariate analysis

Posted on 2001-06-22
4
236 Views
Last Modified: 2010-05-18
i have 10k data points[multiple inputs and outputs.] what methods are available to fit these data points? shall i go for subspace based methods?
0
Comment
Question by:sarika1977
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 2
  • 2
4 Comments
 
LVL 4

Accepted Solution

by:
prashant_n_mhatre earned 200 total points
ID: 6220160
You said you have 10000 data points. If your system is linear 500-700 data points are enough. First plot the data...look at trends...select proper data channel which you think captures the dynamic part.

Check if the data is linear.

Linear:

AR (Auto Regressive), ARX(AutoRegressive eXogeneous), ARMAX(AutoRegressive Moving Average eXogeneous), Box-Jenkins(BJ) models are available. There is not much difference among them if you don't bother about the modeling of residuals. You may go with ARX.


Nonlinear:

NARX (Nonlinear ARX), NARMAX (Nonlinear - ARMAX) and so many options are available. But I would suggest you to try out linear model. If it doesn't fit, you may try out bilinear combinations. Other high order models may not be stable.


Subspace Based Methods:

You may try out PLS (Partial Least Squares), PCA  (Principle Component Analysis), N4SID, CVA etc. They definitely give better fit than others but the order of model will blow up. Ultimately it will result in large computation time.


Feel free to get back to me if something is not clear to you..or need explanation or more info on specific topic
0
 

Author Comment

by:sarika1977
ID: 6220168
pls suggest references on n4sid
thanx
0
 
LVL 4

Expert Comment

by:prashant_n_mhatre
ID: 6220279
I guess you are using MATLAB. see

>> help n4sid

I have some documents. Send me mail@

prashantm@www.com

I'll send you.
0
 

Author Comment

by:sarika1977
ID: 6220337
nice documents
0

Featured Post

Announcing the Most Valuable Experts of 2016

MVEs are more concerned with the satisfaction of those they help than with the considerable points they can earn. They are the types of people you feel privileged to call colleagues. Join us in honoring this amazing group of Experts.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Suggested Solutions

Title # Comments Views Activity
Support for Notepad++ (including downloading & installing a plugin) 5 249
VbScript to countdown to New Year's Day 6 94
learn programming 8 95
Notepad++ Question 11 28
Although it can be difficult to imagine, someday your child will have a career of his or her own. He or she will likely start a family, buy a home and start having their own children. So, while being a kid is still extremely important, it’s also …
Computer science students often experience many of the same frustrations when going through their engineering courses. This article presents seven tips I found useful when completing a bachelors and masters degree in computing which I believe may he…

752 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question