PureBytes Links
Trading Reference Links
|
Andrew Peskin writes:
>
>With the following inputs:
>
> OptimalF = 0.5
> WinPercent = 0.35
> MaxLoss = -3
> AvgWin = 4
> AvgLoss = -1
> Depletion = 0.75
>
>I arrive at: R6 = 0.0898289
>
>Which seems a little low.
>Any and all help would be appreciated.
>
Your *inputs* don't seem self-consistent. Refer to the
very same book (_Portfolio_Management_Formulas_), page
number 83. There is a formula in a section called KELLEY,
for the "optimal f" when winners and losers are not all
the same size. Using your notation above, the formula
is:
((1 + abs(AvgWin/AvgLoss)) * WinPercent) - 1
--------------------------------------------- = OptimalF
abs(AvgWin/AvgLoss)
But plugging in your numerical values,
((1 + abs(4/-1)) * 0.35) - 1
----------------------------- = OptimalF
abs(4/-1)
Therefore
(5 * 0.35) - 1
---------------- = OptimalF
4
which means that 0.1875 = OptimalF.
Yet you specified that OptimalF = 0.5 . Since 0.1875
isn't equal to 0.5 (not even close!), there seems to
be a discrepancy between your inputs
(AvgWin , AvgLoss, WinPercent)
and your input
(OptimalF)
You might want to check this.
--
Mark Johnson Silicon Valley, California mark@xxxxxxxxxxxx
"... The world will little note, nor long remember, what we
say here..." -Abraham Lincoln, "The Gettysburg Address"
|