|
PureBytes Links
Trading Reference Links
|
And now for something in a more serious vein . . .
I have Ryan Jones' software, the version that was released last year, so
my comments pertain to that and not the new version just released. (I
have sent away for that but have not received any email back from him,
which I requested him to - By the way I am still waiting for him to send
me the Dec 97 newsletter so I am not surprised he has not returned your
email either . . . Ryan if you are listening, your customer service is
woeful.)
Money management in this sense is working out how many contracts to
trade as your account equity changes. Research suggests that given a
system with a positive mathematical expectation and that you are a
profit maximiser, you should use an anti-martingale approach ie as your
account equity grows increase the number of contracts you trade. How to
do?
Optimal f: read Ralph Vince's books - brilliant
Ryan Jones' Fixed Ratio: The concept behind Fixed Ratio is excellent and
for most people is going to be more tradeable than say optimal f (cos
your net profit/ maxdrawdown ratio will stay relatively constant for
different levels of delta (read aggressiveness).
Optimal f boils down to having a constant equity per unit (EPU) eg for
every $10K trade 1 contract
My conceptual understanding is that FR requires an increasing EPU eg 1
contract 10K, 2 contracts EPU= $15K, 3 contracts EPU = $20K. This is
what Ryan's algorithm is all about, and if you are still with me note
also that EPU (as a %)increases at a decreasing rate (Pretty neat).
What this means is that as our account grows we still increase contracts
but not as quickly as say optimal f.
a very simplified (may not be entirely accurate but serves the general
purpose) example:
eg 1 contract ----> 2 contracts ----> 3 contracts
Optimal f EPU 10K 10K 10K
Opt f account bal 10K 20K 30K
FR EPU 10K 15K 20K
FR account bal 10K 30K 60K
Notice then that when "account bal" reaches 30K under Opt f we are
trading 3 contracts but only 2 contracts with FR ( because of increasing
EPU).
I think that by requiring increasing EPU, FR mathematically captures
mine and most people's risk aversion in the absolute sense (optimal f
doesn't take this into account and that's why it is so hard too trade).
Ryan's approach is quite elegant in this respect.
eg if my account bal is $10K and I suffer a 50% drawdown that takes me
to $5K. That I think I can handle.
but consider if my account is $1M then that same % DD would take me to
$500K. Dropping half a mill is gonna drive me nuts.
Yes opt f will make you more money but the absolute drawdowns will be
hard to take (vicious circle here). FR seems to me a more balanced
practical approach between profit maximising and risk aversion.
And now back to the software . . .
I don't think it is necessary to purchase the software to use FR. First
go to Ryan and buy all the issues of his newsletter or go to the Futures
Learning Centre and get a video or audio tape. Once you go through the
math, you can set up a spreadsheet that can do all the calcs (assuming
you know your way around a spreadsheet).
For portfolios it will be more difficult but still possible. I am
currently not big on portfolios because the key thing is the expected
correlation between markets/systems is likely non-stationary (yeah just
like your system's returns and price changes). If that is true and you
add money management on top of it, you will likely blow yourself up.
According to Ralph Vince all correlations reduce to 1.0 and so portfolio
diversification (trading more than 1 market because of perceived
negative correlations) is essentially the same as trading multiple
contracts in a single market. The degree to which that is true will
mean that with money mangement you will be trading multiple contracts on
mulitple contracts in a single market - pheeewww - you had better be
sure your historical correlations and back testing is soundly carried
out before doing such a thing or ouuuuch . . .but . . . .
Last but most . . . the math of money management isn't too difficult to
apply. The key assumption is that you have a WINNING SYSTEM IN THE
FUTURE - but you won't know that until after the fact, crystal ball
please - backtest thoroughly but just know that that historical sequence
of trades won't be repeated ever (well maybe not ever). From where I am
standing it pays to be more than a little conservative
Keep moving with the moving target . . .
Peter
Butch wrote:
>
> Van & Timothy, I to am interested in finding a good money management
> package. I have been researching this for a few months. I have
> narrowed it down to:
>
> Rina Systems http://www.rinasystems.com/
> &
> Ryan Jones http://www.smarttrading.com/
>
> Both products are +/- $1600.
>
> I have received a demo of the Rina software & I really like it. It is
> somewhat tied to TradeStation, such that in order to test money
> management strategies your system must be programmable into
> "EasyLanguage". I asked Leo Zimansky (Principal) at Rina about this &
> he said if your system involves discretionary methods you can just tell
> TradeStation where you bought & sold, to accomplish the same thing! I
> use SuperCharts so I have not figured out if I can use their product.
>
> I am really looking for a standalone product. According to Ryan Jones
> web site
> (see below) Performance I may be what I am looking for. I e-mailed them
> a week ago to inquire about a Performance I demo & a video that was
> recommended, I have not received a response. If anyone has some
> information I sure would be interested!
>
> "Performance I is the only software of its kind. Stand-alone Windows
> based 32 bit money management program is also compatible with SuperChart
> and/or TradeStation ($1,699.00)
>
> Butch Barham
> Dallas, TX
|