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

Re: [amibroker] Re: PositionSize / Capital



PureBytes Links

Trading Reference Links

Hi Christoper,

Sunday, December 12, 2004, 11:02:09 AM, you wrote:

C> Jumping in here to make one observation...

C> I've seen much higher returns as well when NOT using volatility based
C> position sizes.  But other stats/metrics really take a plunge.

C> For example - CAR triples while Sharpe becomes 1/5th of previous value
C> and K-Ratio drops to 0.04, from 0.10.  Other things like, Max Sys DD
C> doesn't change that much...

C> That's just my expereince... do other people see these type of results
C> as well?  How much weight can we place on K-Ratio and drop in Sharpe?

It really depends on your Aggressive/Defensive posture.  Personally,
I want to be *very* risk averse.  I am far less concerned about the
positive stats than the negative stats, so I place far more weight on
draw downs than on K- or Sharpe ratios.  I also place a lot of weight
on percentage winners, and consecutive losers, wanting the former to
be high and the latter to be low of course.  One can look at a
system's stats on paper and say, 'this is good, I'll trade this', but
loser after loser can dissuade all but the most (over?)confident. ^_^
After all, it's not a mathematical edge like the house enjoys in
roulette, but an apparently statistically valid edge resulting from
apparently repetitive human behavior.  But math is pretty set in
stone; human behavior is subject to change.

The reason I want to be so risk averse is that I think I can detect
such a behavioral change before I'm bailing water on the Titanic, the
more I require the negative statistics to be absolutely the best
possible. The positive tends to take care of itself.

You don't quantify "that much", but if you are getting three times
the CAR and the DDs are such that, should they increase by 50 to 100
percent, you could live with them, I'd be tempted to take the extra
return.  But the key is remembering that the DDs you are getting are
the worst for the sample period, and not likely to be the worst you
will ever experience (the worst is always just over the horizon).  So
a "that much" extra DD might become a "that much" extra + 50 to 100
percent DD.  If that would still within your psychological ability to
stay on system, I'd see little problem.

But, if the return was great with the lower DD, versus super with the
higher DD, I'd probably settle for great.  To be specific, if I was
expecting, say, 35 percent annualized (let's face it, that is very
serious and rapid wealth creation if it can be continued and
compounded) with very low DDs, I wouldn't be too interested in
shooting for much more unless it came bundled with a *very* small
increase in risk.

But no one statistic is going to decide the matter, or at least it
sure should not.  You have to decide what trade offs you are willing,
and able, to live with.  Different people are going to arrive at
different conclusions looking at the same results.

Yuki



------------------------ Yahoo! Groups Sponsor --------------------~--> 
$4.98 domain names from Yahoo!. Register anything.
http://us.click.yahoo.com/Q7_YsB/neXJAA/yQLSAA/GHeqlB/TM
--------------------------------------------------------------------~-> 

Check AmiBroker web page at:
http://www.amibroker.com/

Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/amibroker/

<*> To unsubscribe from this group, send an email to:
    amibroker-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/