**Date:** Thu, 12 Jan 2006 16:32:49 -0800
**Reply-To:** David L Cassell <davidlcassell@MSN.COM>
**Sender:** "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
**From:** David L Cassell <davidlcassell@MSN.COM>
**Subject:** Re: Monte carlo on Flutter Prediction Process.
**In-Reply-To:** <200601121205.k0CBk7Ma031956@mailgw.cc.uga.edu>
**Content-Type:** text/plain; format=flowed
asimaliabbasi@GMAIL.COM wrote:
>I have been asked to apply Monte Carlo on Flutter Prediction Process,
>but I dont know much about it. Can any body help me in this regard?

Well, I'm sure you were given a bit more direction than what you say above.
But the basic idea is one of repeated sampling from a data set and using
that
set of replicates to run through the process and see what sort of
variability
one can expect.

I can't be more specific until you write back to SAS-L and explain *exactly*
what you are being asked to do.

One use of Monte Carlo methods that you have probably seen without realizing
it is computing the area under the curve. Instead of working with the
integral
of a function y=f(x), you set up a rectangle (x from a to b, where these are
the
bounds for the problem; y from, say, 0 to something more than the max of the
curve in this range - let's call it M). So the box has area (b-a)*(M-0).
Now
throw 10,000 random points in there. How many of them are below the line?
7,348 you tell me. Then our estimate of the area under the curve from a to
b
is

area = (b-a)*(M-0) * 7348/10000

Now you'll have a better idea of what this is, and you can think about what
the
goal of the Monte Carlo simulation really is. When you write back, I'll
probably
specify something using PROC SURVEYSELECT and by-processing. I'm really
predictable.

HTH,
David
--
David L. Cassell
mathematical statistician
Design Pathways
3115 NW Norwood Pl.
Corvallis OR 97330

_________________________________________________________________
On the road to retirement? Check out MSN Life Events for advice on how to
get there! http://lifeevents.msn.com/category.aspx?cid=Retirement