Re: Polyfit with three variable

[Andreas Stahel]

mmuetzel wrote
> Am 20. Dezember 2020 um 15:57 Uhr schrieb "swagata Mukherjee":
>> Hi,
>> My question is whether we have any method by which I can use regression
>> with 3 variables like polyfit(x,y,z,N)?
>> Regards 
> 
> This answer on the discourse forum might be helpful:
> https://octave.discourse.group/t/execucao-do-codigo/497/8
> 
> In principle, that approach also works for multi-dimensional polynomials
> of higher degree.
> 
> HTH,
> Markus

The linl given by Markus is a good one.

You might have a look at the command LinearRegression() in the optimization
package or the command regress().
To deal with multiple variable have a look at Section 2.2 in
https://web.sha1.bfh.science/Labs/PWF/Documentation/OctaveAtBFH.pdf
Example 2 (p. 162) and Example 4 (p. 168) use multiple variables.

I hope this helps

Andreas




--
Sent from: https://octave.1599824.n4.nabble.com/Octave-General-f1599825.html