[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fixed Ratio Function



PureBytes Links

Trading Reference Links

Congratulations to Ken Lindauer for applying
the Quadratic Formula to get the closed form
expression for #contracts = f(equity).  Also
congratulations for making the more subtle observation
that Jones's formula depends, in fact, upon
NetProfits *only*, and not [for example] on
the amount of starting equity in the account.
Thus we have the paradoxical result that if
two traders, Alice and Bob, start trading
some system like Oddball at the same time,
but Alice starts with $300,000 and Bob starts
with $30,000, the given formula will have them
trade the exact same number of contracts,
on the first trade (when Alice has 10X more
money than Bob) and on all subsequent trades.

However, please remember that this is only the
easy part.  This is the part that Jones calls
"Rate of Increase" or ROI.  The *hard* part is the
part that Jones calls "Rate of Decrease" or ROD.

The reason why ROD is difficult is that you need
access to the entire portfolio-level equity history
stretching back to the first day of trading.
Unlike ROI, ROD is "path dependent" -- what you
do today is a function of today's equity AND of
the historical events that got you to this equity.

Behold! doesn't provide this.  Trading Recipes doesn't
provide this.  Athena doesn't provide this.

You can fudge it in Tradestation, as long as you
are willing to have a "portfolio" containing only
one tradeable and only one tradingsystem.

So, as a result, most folks either pretend they
never read those chapters of Jones's book, or they
do some kind of roll-your-own specialized software
to handle ROD.  Jones himself used to sell a standalone
program called "Performance I" which did perform the
ROD calculations.

Here's a little challenge for the math enthusiasts
on the list.  Jones says that he recommends switching
from Fixed Ratio to Fixed Fractional once the account
has grown in size.  If this is a good idea, then
there ought to be mathematical solutions to these
problems:
(1) invent a criterion whereby you can judge whether
Fixed Ratio or Fixed Fractional is preferable, at any
particular time.
(2) using your criterion, find the OPTIMUM point at
which it is best to switch from FR to FF.

Perhaps Jones's book already answers these problems;
if so, I haven't been able to find those passages.
--

   Mark Johnson     Silicon Valley, California     mark@xxxxxxxxxxxx

   "... The world will little note, nor long remember, what we
    say here..."   -Abraham Lincoln, "The Gettysburg Address"