[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
## Re: [Help-gsl] linear regression, f-statistic and vectors

**From**: |
Brian Gough |

**Subject**: |
Re: [Help-gsl] linear regression, f-statistic and vectors |

**Date**: |
Sat, 28 Jul 2007 20:16:14 +0100 |

**User-agent**: |
Wanderlust/2.14.0 (Africa) Emacs/22.1 Mule/5.0 (SAKAKI) |

At Thu, 26 Jul 2007 12:53:36 +0200,
Benjamin Otto wrote:
>* gsl_fit_linear expects two array of type double so for my previous tests*
>* I just created some matching arrays. For future calculations I would*
>* like to use vectors. Now can I transform a gsl_vector into such a*
>* "normal" double array so it is accepted by the linear regression*
>* function? I don't find any matching vector view in the manual. Or do I*
>* have to loop over each single element, get the value and copy it into a*
>* new double array?*
The gsl vector types are mainly intended for linear algebra. If you
just have "data" it's simpler to use an ordinary array for it -- this
is why the statistics functions only take arrays.
We don't have a routine for the f-test, but you can obtain the
F-distribution parameters from the cdf functions.
--
Brian Gough