|
PureBytes Links
Trading Reference Links
|
If you deal with complex issues like real portfolio money management you
will soon realize that the starting date can influence the result
dramatically. Also the position size that you pick for your trading is
important. You could read the system lab articles of the Active Trader. In
some I am showing the dramatic effect of it. If you do not use real
portfolio level backtesting methods in your money management settings then
you should consider doing that. Finally I would recommend MonteCarlo
simulation of the portfolio results. This should give you a good idea of
what to expect...
VK
-----Original Message-----
From: Frank Fleisher [mailto:r4_6fpen8@xxxxxxxxxxxxx]
Sent: Thursday, June 17, 2004 5:59 PM
To: omega-list@xxxxxxxxxx
Subject: Re: Equity complexity
Hello Cameron,
The number of systems and markets is arbitrary, but if you find
yourself with a very large portfolio, and/or systems that generate frequent
trades, you
must have a realistic assessment of slippage caused by the lack of
transparency/liquidity in your chosen markets, and put it in your
analysis.
You want a least-correlated combination of markets and
trading systems, and work with percent equity curves. In your equity curve
analysis, don't focus on the
equity curve's random walk as long as it walks within a statistical
range that is comfortable to you and points up. So, for example,
don't "chase the curve" and think about doubling your investment
just because the curve moves at a higher rate than the normal. You should
increase your investment if your comfortable with the corresponding $$
volatility it
brings. Finally, don't rely on an equity curve built on
a hand full of trades going back a few months.
"Beyond Technical Analysis," by Tushar S. Chande, has a
straight-forward chapter on equity curve analysis.
-F
Thursday, June 17, 2004, 5:15:54 AM, you wrote:
c> Hi all
c> I am analysing portfolios and , i am wondering..... What criteria do i
pick
c> to choose which systems and securities to trade?
c> I was thinking the best ROA [return on account,net profit/max drawdown]
but
c> then i though smallest flat period of the final equity curve........ but
c> after analysing these at what point would you consider to many
c> systems/securities? Does inputing orders on 15 markets sound ummmmmmmm
over
c> the top ? what about when my equity increases do i add more
securities....
c> what about 20 then 60 then 150 ? when is enough enough?
c> or.... do i pick the best 5 systems and then start doubling the
contracts
c> traded which also doubles the drawdowns
c> Any feedback welcome :O)
c> Any books recommended for me to read?
c> Cheers
c> Cameron
--
Best regards,
Frank mailto:r4_6fpen8@xxxxxxxxxxxxx
|