[EMAIL PROTECTED] (Kristian Moe) wrote:

> [...] I'm working on a finance thesis, focusing on 
> options and value at risk.

Given this, the error of the interpolation scheme probably
won't have any noticeable effect. The numbers you quote seem
to be some empirical results -- if so, the error introduced
by using the wrong model (e.g., assuming independence and/or
Gaussian distribution of some important quantity) are going
to greatly outweigh the error of the interpolation scheme. 

If there is some theory about the distribution of the results
(e.g., results should be Poisson distributed or something)
then, by all means, determine the parameters of a theoretical
distribution and integrate w.r.t. that distribution. This is
equivalent to implementing an interpolation scheme based on
the theory. Failing that, linear interpolation will likely
work as well as anything. 

It seems likely that you've presented only a small part of
a large, broad problem. My advice is that you should work on
covering all parts of the problem in a reasonable way, rather
than spending a lot of time on an unimportant detail at the
expense of other parts. 

For what it's worth,
Robert Dodier
--
Far better an approximate answer to the right question, which is often
vague, than an exact answer to the wrong question, which can always be
made precise. -- John W. Tukey
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to