Re: detrend() in Matlab and Octave

[Sergei Steshenko]

--- On Wed, 12/22/10, Ben Abbott <address@hidden> wrote:

> From: Ben Abbott <address@hidden>
> Subject: Re: detrend() in Matlab and Octave
> To: "H W Borchers" <address@hidden>
> Cc: address@hidden
> Date: Wednesday, December 22, 2010, 3:12 PM
> On Dec 22, 2010, at 1:23 PM, H W
> Borchers wrote:
> 
> > The 'detrend()' functions in Matlab and Octave provide
> quite different
> > functionality:
> > 
> > Matlab: Removes a constant or linear trend between
> given breakpoints.
> > 
> > Octave: Removes a trend of order n from the whole
> sequence.
> > 
> > What is the reason for this varied behavior?
> > 
> > Hans Werner
> 
> I'm not sure I understand the context of your question. But
> I'll take a shot. I'll assume you are referring to the
> documented features of Matlab's implementation of detrend
> that Octave does not support?
> 
> Compatibility with Matlab wasn't always a high priority for
> Octave development. As a result, there remain some functions
> that are not fully compatible.
> 
> The most likely reason Octave's detrend is not fully
> compatible is because no one is aware of the problem, or no
> one has invested the effort to fix it.
> 
> If you'd like to see the compatibility of detrend fixed,
> please file a bug report ...
> 
>     https://savannah.gnu.org/bugs/?group=octave
> 
> Or better yet, If you are particularly motived, please
> submit a patch.
> 
> Ben
> 

The existing Octave 'detrend' should then remain with a different name -
because it has features the Matlab one doesn't have.

Regards,
  Sergei.