comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » Re: Regression fit and random noise
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Return to the default flat view Create a new topic Submit Reply
Re: Regression fit and random noise [message #83687 is a reply to message #83686] Thu, 28 March 2013 16:26 Go to previous messageGo to previous message
Mats Löfdahl is currently offline  Mats Löfdahl
Messages: 263
Registered: January 2012
Senior Member
Den torsdagen den 28:e mars 2013 kl. 22:26:01 UTC skrev kisCA:
> Thank you Phillip for your answer and the book reference !
>
> What I am trying to know is if by increasing the noise ratio in my data, the model will still find a good fit.
>
> I increase the noise with:
>
> new_sig = original_sig + noise_ratio*randomu(seed,n_elements(original_sig))
>
> I increase slowly the value of noise_ratio. So first I obtain almost the same value of R2 as if no noise was there. R2 is getting lower as the value of noise_ratio increase. After a certain value of noise_ratio is reached, R2 values don't get lower than 0.3.

Maybe I misunderstand what you are trying to do but... Are you aware that randomu has a uniform distribution between 0 and 1? So you are adding on the average something like 0.5*noise_ratio to your original signal. So maybe you want to add noise_ratio*(randomu(...)-0.5) instead. Or, since randomn is normal distributed with zero mean, simply noise_ratio*randomn(...).
[Message index]
 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic: Simplify if statements and for loops inside for loops
Next Topic: Re: Calculating difference between each element in array from previous

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Sat Oct 11 01:35:37 PDT 2025

Total time taken to generate the page: 0.87919 seconds