Re: 8 independent variable curve fitting.

[Olaf Till]

On Thu, Jun 04, 2015 at 10:37:37AM -0400, Doug Stewart wrote:
> <snip>
> To help me understand how to use leasqr I have made up a demo.
> See the attached file.
> This file has more in it than just a demo of leasqr. It has comments about
> the general idea of multi-variable regression.
> 
> I think it should be adapted for a demo of ???? (leasqr  , multi-variable
> regression other functions). The leasqr could be replaced by other curve
> fitting functions.
> Olaf and Jaun  what do you think.
> Doug

Doug,

by reading your demo I'm able to answer my own previous question:

On Tue, Jun 02, 2015 at 12:36:28PM +0200, Olaf Till wrote:
> Not sure if I understand you right. What are the variables you want to
> determine by fitting? What variables do you assume to be given in the
> data points? What are the variables in which your relation is
> non-linear, the variables you want to determine by fitting or other
> variables?

but it would have been more efficient for you to answer to it.

What you demonstrate is linear regression. It plays no role that your
equations are non-linear in 'x'. What matters is that they are linear
in 'p'.

Olaf

-- 
public key id EAFE0591, e.g. on x-hkp://pool.sks-keyservers.net

signature.asc
Description: Digital signature