[R] fitting distributions

[Prof Brian Ripley]

On Sat, 30 Jun 2001, osama wrote:

> Hi;
>     I would like to fit a contiuous right skewed dist. to a set of data.
So, I want to use MLE or Robust estimation of the parameters, then use
A-D or K-S test or Robust alternative for the goodness-of -fit. Also I
would rank the fitted distributions according to the p-value of the test.
Does anyone know how I do this in R. any backages of Robust estimation
and for goodness-of-fit?

Is this data analysis or a homework problem?

What robust estimation even means for a right skewed distribution is
problematical.  The standard methods for the mean assume symmetry.
As robustness menas considering distributions outside the specified
family, you can only robustly estimate parameters with a physical meaning.

An in general the p-values of A-D or K-S tests assume known parameters.
(Yes, I know Stephens and others have worked on approximations in other
cases.)

R provides methods for real data analysis, and I don't see why one would
need to fit lots of (presumably) univariate distributions.  Find a good
density estimate instead.  But theoretical statistics do set this sort of
thing as problems ....


> Thank you for your help.
> Ahmed hassan
> alexandria univeristy
>
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._