|
PureBytes Links
Trading Reference Links
|
Dave,
My comments are below.
--- In amibroker@xxxxxxxxxxxxxxx, "Dave Merrill" <dmerrill@xxxx> wrote:
> Mark, thanks a ton for doing this
You're welcome.
>
> > Test *unoptimized* system
>
> hmmm. what does unoptimized mean? "traditional" parameter settings,
if they
> exist, like MACD(12, 26, 9)? it's hard to see how there is such a
thing as
> "neutral" settings. the traditional settings must have come from
tests on
> some particular type of equity in some specific time frame. but I
understand
> you're saying not to tweak for this portfolio and time frame.
I start with the settings someone provides or default settings and
expand from there if I think there's potential (as outlined in my post
to Dale). But don't limit yourself to traditional indicators. For
example, look at patterns consisting of relationships between price
arrays for different numbers of bars.
> questions:
>
> - I'm not coming up with a plausable 2 year sidewise period. I see
periods
> that long that net flat, but they have significant rises and falls
within
> them, which isn't really the idea. I know you don't want to discuss your
> proprietary time periods in too much detail, but I'm looking at the naz
> composite; is that the problem? got a quantified definition of "flat"?
>
Think in terms of blunt tools. Nothing magic about 2 year intervals
or any definition of flat that I can think of. Looking at a chart,
there are recent 2 year bull and bear periods that are pretty obvious.
If you don't see a 2 year flat period you're comfortable with, choose
a smaller window. Bottom line is pretty much no matter what you
choose, if you find systems that are profitable across the stocks and
time periods as I've outlined, they're probably robust.
> - do you use the current components of each index for all your
tests? or do
> you have historical versions of each of them, corresponding to the time
> periods you test with? if so, where do you get them?
Current.
Regards,
Mark
>
> -----------
>
> not directly related to this specifc post, but relevant, I've been
thinking
> about trading systems as dividing into two basic camps, tracking
systems and
> opportunistic systems. tracking systems try to call the profitable
market
> direction at any given time for each equity you point them at.
opportunistic
> systems look for profitable setups somewhere in the universe of equities
> they're watching, and may have no opinion on any particular equity
at any
> particular time.
>
> these two methods are fundamentally different. one needs to
understand the
> dynamics behind each stock well enough to try to sense its
forthcoming price
> moves all the time. opportunistic systems may not understand
anything at all
> about 98% of the price action on each stock, profiting instead from
the 2%
> of the time when they recognize a pattern whose implications about the
> future they can predict.
>
> do we need to test these two types of systems differently? an
opportunistic
> system knows when it works and when it doesn't, and sticks to what
it knows.
> tracking systems need to function in all kinds of weather to be
successful.
>
> dave
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->
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
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|