Edgeworth Binomial Model

JimKallimani · November 21st, 2003, 11:43 am

So what do you guys think of it?

AlanB · November 21st, 2003, 11:46 am

I've not heard of it - what is it? Is there any literature on it?

JimKallimani · November 21st, 2003, 11:49 am

http://www.haas.berkeley.edu/finance/WP/rpf275.pdf by Mark Rubinstein.

AlanB · November 21st, 2003, 11:59 am

Actually, the Edgeworth expansion is used by Turnbull & Wakeman to value Asian options. When compared to the exact solution to Asian option valuation by Linetsky and separately by Lewis, the Edgeworth expansion is found not to be sufficient, in certain limits. That certainly could be a function of the Asian option itself. It may work better for other derivatives, but I'd be careful to examine the limitations. You can use Google to find the articles by the authors I've referenced above.

apine · November 21st, 2003, 8:23 pm

speaking for myself, the problem with that (and gram-charlier) is that for "extreme" skew & kurtosis combinations, it doesn't give a good distribution. as the "extreme" is not very extreme at all, it makes it difficult to use. my guess is that others feel similarly, but i would like to hear others, too.