[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-gsl] [Bug-gsl] Doubt (bug?) in gsl_fit_linear function
From: |
Juan Pablo Amorocho D. |
Subject: |
Re: [Help-gsl] [Bug-gsl] Doubt (bug?) in gsl_fit_linear function |
Date: |
Mon, 11 Jun 2012 10:51:26 +0200 |
@Frank. Thank you. That's a sound answer!
On Jun 11, 2012 10:35 AM, "Frank Reininghaus" <address@hidden>
wrote:
> Hi,
>
> 2012/6/11 Juan Pablo Amorocho D.:
> > What should be (computationally) the value of the intercept in this case?
> > Nan seems right to me, but the slope /should/ be inf. Any comments?
>
> why should the slope be inf? It could just as well be -inf, couldn't
> it? In other words: the slope is undefined, in which case returning
> NaN is the most sensible thing to do IMHO.
>
> In any case, you wouldn't be able to use the slope for anything useful
> later on. Note that gsl_fit_linear() tries to fit data with a function
> y=m*x+b, which makes no sense at all for data points on a vertical
> line.
>
> Cheers,
> Frank
>