Is there a definite need for fitting the distribution? If you have to
try many distributions or fit many parameters, the mean squared errors of
the resulting distribution estimates are no lower than that of
nonparametric estimates (empirical CDF or kernel density estimator).
Frank Harrell
On Wed, 02 Jan 2002 10:48:12 -0500, Chia C Chong wrote:
> Hi!!
>
> I have a set of data with some kind of distribution. When I plotted the
> histogram density of this set of data, it looks sth like the
> Weibull/Exp/Gamma distribution. I find the parameters that best fit the
> data and then, plot the respective distribution using the estimated
> parameters on the empirical distribution. My question is, what kind of
> statistical test that I should use so that I will know which estimated
> distribution will fit the data better?? I need some kind of test that
> will give me some numerical values which distribution is fit better
> rather than just observed the fitting graphically..
>
> Thanks for th help in advance..
>
> Regards,
> CCC
>
>
>
>
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at
http://jse.stat.ncsu.edu/
=================================================================
