PureBytes Links
Trading Reference Links
|
As they say ... don't confuse brilliance with a bull market ... pre
3/2000 for a couple of years one didn't need much more than a
newspaper and some darts. Personally I walk forward test all
systems ... They either work or they don't out of sample. Pre 2000
one had to look signifigantly farther back for bears ... '87 doesn't
really qualify.
--- In amibroker@xxxxxxxxxxxxxxx, "seneca_kw" <seneca_kw@xxxx> wrote:
>
> Fred,
>
> In my testing its commonplace to find setups that work very well up
> to year 2000, then completely fall apart. It's almost a routine
> now. If a 10 year test returns encouraging results, I then test
year
> by year. If the first years start out gangbusters I know exactly
> what that means: the most recent years are killers.
>
> Wayne
>
>
> --- In amibroker@xxxxxxxxxxxxxxx, "Fred" <ftonetti@xxxx> wrote:
> >
> > It's too bad this study apparently ended in 2000 as it would have
> > been interesting to see the results since then forward as well.
> >
> > --- In amibroker@xxxxxxxxxxxxxxx, "seneca_kw" <seneca_kw@xxxx>
> wrote:
> > >
> > > Duke,
> > >
> > > Thanks for the interesting link. I hadn't seen that study
> > before.
> > > It shows that a combination of TA and FA can be successful, but
> it
> > > doesn't quite answer the question that I had in mind.
> > >
> > > Take the example of a simple reversion-to-the-mean system: buy
> > when a
> > > stock closes below the lower Bollinger Band and exit N days
> > later.
> > > Does adding a fundamentals screen help? To test this, I'd
divide
> > > stocks into at least five categories, from the lowest-rated
> > > fundamentals to the highest. Then I'd test each category using
> > the
> > > same system paramenters. Ideally, the results should be worst
> for
> > > the lowest-rated fundamentals, and should improve uniformly and
> > > consistently up to the highest-rated. That would show that
using
> > > fundamentals adds value.
> > >
> > > But even if using fundamentals increases the profit per trade,
it
> > > doesn't necessarily follow that you'd want to incorporate them
> > into
> > > your system. They may decrease the number of signals to the
> point
> > > that your overall profits are lower even though your per-trade
> > profit
> > > is higher. In the example system, I know that I can improve
per-
> > > trade profits by tightening the requirements (eg stock must
close
> > at
> > > 90% of lower BB). Maybe I'm better off chucking the
fundamentals
> > > screen, tightening the BB requirements, and screening the whole
> > > market (which is what I think the original poster was asking).
> > > These are the kinds of questions that I'm interested in
> > investigating.
> > > Wayne
> > >
> > >
> > > --- In amibroker@xxxxxxxxxxxxxxx, "duke.jones"
<Duke.Jones@xxxx>
> > > wrote:
> > > > Wayne,
> > > >
> > > > Here is a PDF from Charlie Kirkpatrick which discusses a real
> > time
> > > portfolio using just three elements. Two of which are
fundamental
> > the
> > > third price momentum.
> > http://www.mta.org/awards/01/2001DowAwardb.pdf
> > > >
> > > > I believe fundamentals can be used to increase the
probability
> > of
> > > success (based on testing and results) but the key is how you
> > measure
> > > success. Kirkpatrick's strategy has continued to perform well
and
> > has
> > > consistently beaten the market but you had better be able to
> > stomach
> > > the large drawdowns. I have a enclosed pic of real time
> > performance
> > > since the beginning of last year of the Kirkpatrick (kirk.gif)
> > model.
> > > As you can see relative performance is great but its a model
that
> > > needs a trending market. Also enclosed is a backtest of a
> > modified
> > > version (valuemo.gif) with more history. Better equity curve
and
> > > roughly half the risk of the market but still large drawdowns.
> > > >
> > > > Where I have found value is using a combination of systems
with
> > > little multicollinearity. I would to love tell you its made me
> > rich
> > > beyond my wildest dreams and that I only post here for the
> > > intellectual curiosity however, the reality is like all systems
> > mine
> > > is a work in progress. The good news is that in aggreagte they
do
> > > have an equity curve I can live with and actually trade. Since
my
> > > primary job is to provide research I also like the fact that
you
> > > don't hear about too many fund/tech systems so perhaps where
> there
> > is
> > > no crowd there is more opportunity.
> > > >
> > > > OK, I have beaten the horse dead..time to climb back into the
> > > shadows.
> > > >
> > > >
> > > > Duke Jones, CMT
------------------------ Yahoo! Groups Sponsor --------------------~-->
What would our lives be like without music, dance, and theater?
Donate or volunteer in the arts today at Network for Good!
http://us.click.yahoo.com/Tcy2bD/SOnJAA/cosFAA/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/
|