A couple of approaches would be an Edgeworth expansion
http://www.math.sfu.ca/~cbm/aands/page_935.htm
or the so called "Skew Normal" distribution
http://tango.stat.unipd.it/SN/

Jerry

"James Silverton" wrote:

Hello, Mike!
You wrote on Sun, 17 Dec 2006 22:12:15 -0800:

MM The normal curve (bell-shaped) is completely described by
MM only two parameters: the mean and standard deviaton.

MM For charting instructions, see

MM
http://www.tushar-mehta.com/excel/ch...tion/index.htm

MM - Mike
MM http://www.mikemiddleton.com

This is doubtless correct but I suppose you could modify the
values given by a normal curve to show kurtosis and skewness by
using a suitable non-linear scale. Still, I'm not sure how I
would achieve a given value of either.

MM "MarkS" wrote in message
MM ...
?? Hi,
?? I know the mean, max, min, standard deviation, skew and
?? kurtosis.
??
?? I need to plot a graph with this data and I have got as
?? far as using NORMDIST to create the normal distribution
?? curve. How do I add in the skew and kurtosis
??
?? Thanks


James Silverton
Potomac, Maryland

E-mail, with obvious alterations:
not.jim.silverton.at.comcast.not