|
PureBytes Links
Trading Reference Links
|
Howard and Peter,
Thank you for the replies. They are helping me sort through
the concepts of testing stops, especially as they relate to
systems that use market timing.
Howard, you were right on when you state the value of exits
depends in part on whether new trade entries are possible.
My system generates good entries for only 1 to 3 days after
a market turn signal is received. New entries are not
available after a sustained market trend is underway and it
is those longer trends that make the system profitable.
A light bulb just went on -- For a system that gives
limited entry points (such as 1-3 days after a market
turn), exits need to be evaluated IN CONJUNCTION WITH the
entry. That seems clear to me, at least for the type of
system I am thinking of.
I wonder if systems that are primarily based on market
timing signals are a especial case. It seems that good
exits from random entries will tell very little about what
is good for many market timing systems.
Let me explain. I will give as much detail as I am
comfortable with. You will recognize that, unlike you, I
operate the belief that my system would degrade if its
details were know (due to its 1-3 day window of opportunity
and its rather narrow filters for stocks - out of thousands
of stocks there are not always enough candidates to give 10
trades on the 1st day of a market signal - hence the window
is up to 3 days). So forgive me if some parts of the
following do not give precise details.
And I have 3 distinct exits for that system: a stop loss
(to catch bad entries), a profit target stop (which is
designed to give optimal exits for a minority of the stocks
which do substantial better than the average trade), and
finally a market timing exit (which is the exit for most
trades). I already know that the optimal (which includes
robust dimension) settings for the stop loss and profit
stops are very broad compared to what would appear to be
optimal with random entries.
Learning the best exits for random entries would, for me,
just be an academic exercise. Of course, my focus is on
systems based on market timing. Results for tests based on
random entries might be very helpful for systems that do
not use market timing in the manner I do.
Thanks again for sharing your thoughts and perspectives.
b
--- pcwinch <pcwinch@xxxxxxxxxxxxxxx> wrote:
> B,
>
> Do not agonise...it may sort itself out. Exiting earlier
> may reduce drawdowns, you may be able to shift more money
> into better trades elsewhere. I do not see any
> intuitiveness here in what you say, and the only way to
> find out is to do it. We could debate till the cows come
> home, but easiest just to try and get it to work. If you
> get out of a trade that later goes onwards and upwards,
> oh well that is life, but your money presumably went into
> something else. Just be zen, nobody expects to get the
> bottoms and tops anyway. It might be surprising how much
> better your overall outcome could be even if a small
> reduction in DD is made because of the benefit of
> compounding..
>
> OK?
>
> Peter
__________________________________
Do you Yahoo!?
Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
http://hotjobs.sweepstakes.yahoo.com/signingbonus
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:
http://docs.yahoo.com/info/terms/
|