Question on min variance hedge EditGrid....

[sridhar]

http://www.bionicturtle.com/premium/editgrid/2008_frm_hull_derivatives_minimum_variance_hedge/

In re the above (I am looking at the XLS version...)

You are trying to illustrate the efficacy of the hedge by wondering what happens if the jet price increases by $1.00.

You compute that the futures value will increase by: 1 / (correlation coeff)^2

What is the basis for this computation? I have not seen this relationship before....Can you elucidate?

--sridhar

 

[David Harper CFA FRM]

sridhar,

It's my brain fart, i tried to use a too-cute shortcut i have seen, but you are not the first to notice this (be confused by) error.

I just modified the min variance XLS to correct the mistake. If you re-look at the bottom, I conduct a proper (manual) test of the hedge...

Now, to "test" the hedge, I use the slope of the regression line (which is the same as the optimal hedge ratio). The R^2 was working by sheer coincedence...Thanks!

David