[R] Distribution transformations

kjetil@entelnet.bo kjetil at entelnet.bo
Sun Nov 23 19:54:24 CET 2003

On 23 Nov 2003 at 19:35, Viola Rossini wrote:

The frechet dirtribution is in the evd (extreme value dist) package
on CRAN.

The basic preinciple is that if U is uniform (0,1) anf F
is a cumulative distrubution function, then
F^{-1}(U) is distributed as F.

Kjetil Halvorsen

> I am still not getting it.
> I am trying to understand multivariate distributions and copulas. In
> the beginning of each article it is said that the observed data must
> be transformed to uniform or frechet distribution by means of
> probability integral transform. Apparently this is something easy and
> trivial and a standard procedure in introductory statistics. Well, I
> have some books in statistics of different degrees of complexity but
> unfortunately I cannot find the answer there. All examples are only
> about how to generate a random sample with desired (always
> exponential) distribution. Now, I have two variables X and Y and I
> want to transform them to Frechet or uniform. I was just thinking, if
> this is so simple and trivial as all stat books say, then, it must
> exist a simple function for it in R.
>
>
> P.s. @Rolf, I would like to have it as a homework but I am afraid I am
> too old for a school.
>
> Rolf Turner <rolf at math.unb.ca> wrote:
> You wrote:
>
> > Dear R-Users,
> >
> > I have a question that bothers me in the last few days. It is
> > supposed to be easy but I can't come up with a solution. Are there
> > any functions in R dealing with transforming empirical and
> > parametric distributions? I have two data sets of observed variables
> > that I want to transform to Frechet and Uniform distribution. I
> > would appreciate if someone could inform me about R-functions for
> > this purpose or enligthen me how to do it by myself.
> >
> > Thank you very much in advance,
> > Viola Rossini
>
> Is this a homework question?
>
> cheers,
>
> Rolf Turner
> rolf at math.unb.ca
>
>
>
> ---------------------------------
>
>  [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help