On May 28, 12:44*pm, Gary''s Student
wrote:
You can STILL get an estimate of the probability by performing
a Monti Carlo analysis.

That's Monte Carlo, with an "e", at least in English.

The first step is to make data. *Pick 5 at random, calculate the average
and store the result. *Repeat this many, many times. *Say we end up
with data from A1 thru A10000 (each cell containing an average)

If you are going to go to the trouble of generating that many 5-tuple
selections, you might as well generate all selections of 5 out of 14.
There are only 2002.

The third step is to add counts to the bins based upon the 10,000
values. *The bins represent a probability distribution.

Or simply sort the 2002 data (averages of 5)....

The last step is to compare the reference value with the bin counts.

....And count the number greater than the reference value.

(If you are going to store the 2002-data sorted table in a worksheet,
you could use LOOKUP.)

But I don't think that is a good general solution. It may or may not
be what the instructor is looking for, depending on what class this is
for.

I will post a separate posting with a more general statistical
approach, if someone else does not beat me to it. I don't have time
at the moment.

(I might even get around to posting the VB function to generate all
2002 combinations. It's quite straight-forward.)