In article <397cfc9a$[EMAIL PROTECTED]>,
Veeral Patel <[EMAIL PROTECTED]> wrote:
>Hi,
>I have a set of data (25000 samples), i have plotted a histogram , the
>historgram is skewed to the right and the closest distribution i can match
>the data to is either gamma or weibull distribution. I would like to
>extrapolate the tail of this pdf. I only have discrete and single (i.e. not
>paired) data. Apart from using the old extrapolative techniques such as EVT,
>Importance Sampling etc are there any new algorithms that i can use?
This is so fraught with danger that, unless you have a
STRONG THEORETICAL reason for a particular form or class
of forms, I would have to advise against even considering
this type of conclusion jumping.
Again, without this type of strong theory, you can do a
fair job of locating the 99.9% point of the distribution,
but there is an 8% chance (before the observations were
taken) that there are no observations past the 99.99%
point, and the 99.999% point is something about which you
can only say it has a very good chance of exceeding your
largest observation, but nothing about putting a
statistical upper bound on it.
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
