|
PureBytes Links
Trading Reference Links
|
Hi b –
I think a lot depends on the time frame of
the trade and the ease of finding new trades.
Any exit methodology (other than a crystal
ball or a future leak in the code) will exit too early or too late. How
much is “too” depends, in my opinion, on the drawdown experienced
by staying in the trade longer or exiting earlier.
Assuming there are lots of opportunities
to take new positions, I prefer to exit early and move on to another trade,
rather than stay too long and not get out until the trade has run its
course.
Thanks,
Howard
-----Original Message-----
From: b [mailto:b519b@xxxxxxxxx]
Sent: Sunday, January
11, 2004 <span
>3:36 PM<font size=2
face=Tahoma>
To: amibroker@xxxxxxxxxxxxxxx
Subject: [amibroker] Exit Study -
A conceptual question
<font size=2
face="Courier New">Howard,
That is
impressive looking code. It will help me pick up
some coding
technique in addition to providing a tool for
evaluating
stops and exits.
I think you
have zeroed in on an often neglected area of
trading -
exits. It is an area where I have so far taken a
very simple
approach: aside from stop loss exits, I usually
rely on
market timing signals for trade exits.
But as I
attempt to set up a test for my market timing
exits I find
myself in a bit of conceptual puzzle. Although
I use some
qualifying filter to improve the efficiency of
the entries,
but these systems work because they stay in
sustained
trends. Any exit before the market trend is over
is
"bad" in the sense that it removes profit potential of
the overall
system. That is because, if a trade is exited
in mid
market trend, my systems do not have a reentry rule.
Without
running any test code I already know the exits
appear to be
"inefficient" on a trade by trade basis but if
I were to
make them more efficient on a trade by trade
basis, the over
all system will suffer (because the overall
system only
enters at market turns). I think I am repeating
myself.
Let me put
this another way, it appears to me that the
value of the
exit for my market trend following systems
lies in the
fact that they allow the market trend entries
to be used.
Thus, I can
not see how I can do any relevant testing of my
exits by
using random entries (or perfectly good or
perfectly
bad entries)? Am I missing something?
Can you, or anyone,
provide a hint how I mgiht get out of
my
conceptual maze?
Thanks in
advance.
b
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
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 the Yahoo! Terms of Service.
|