|
PureBytes Links
Trading Reference Links
|
Herman,
It doesn't answer all of them, but it can answer a lot of them. In
the earlier example Jayson used where we were talking about a MA
length, we obviously won't be cutting that finer then on an integer
basis but if one is looking at an indicator reaching a ceratin level
between 0 and 100 and what we find is that 67 and 68 work well then
we very well might want to look at say 66 to 69 by 0.1 because the
sweet spot might be 66.9 to 68.9 as opposed to 66.0 to 68.0. IMHO
the wider the range the more robust but even when it's not real wide
one wants to be as close to the middle as possible.
Fred
--- In amibroker@xxxxxxxxxxxxxxx, "Herman van den Bergen"
<psytek@xxxx> wrote:
> Hello,
>
> fwiw... there is a lot of eagerness to generalize for the sake of
> convenience. People talk about the markets as if there is only one
way to
> look at them. My experience (whatever that means) is that for some
systems
> the trading principle hasn't failed for over 11 years (limit of my
data) but
> for others I see right-across-the-market drastic changes at some
point in
> time. To me this means that one should assess the market with
respect to the
> systems one is trading.
>
> It is the same with optimizing...some systems give a broad plateau
of
> performance and others give one single very sharp peak. Having a
single
> sharp peak may simply mean you have identified a uniquely stable
market
> characteristic and it should be interpreted in the context of other
> measurements, like the number of winning trades and W/L ratio.
>
> I don't think it is correct to simply generalize and say broad
plateaus are
> good. When optimizing one is also dealing with different types of
variables,
> for some variables one may never be able to look for broad
plateaus, for
> example the periods for a fast trading system; there may only be a
3-4
> values for the period that make any sense. On the other hand,
trigger
> thresholds may have a nice large range and can be plotted nicely on
a
> 100x100 grid - sometimes you may been have to choose between two
high
> performance centers. It would take additional analysis to decide
which one
> to go with. So a 3D plot doesn't answer all questions.
>
> Herman.
>
>
> -----Original Message-----
> From: Fred <fctonetti@xxxx> [mailto:fctonetti@x...]
> Sent: Thursday, February 06, 2003 9:01 AM
> To: amibroker@xxxxxxxxxxxxxxx
> Subject: [amibroker] Nature of the market - Previously Stock
Selection
>
>
> Ken,
>
> This is NOT meant as a comment directed at your statment because I
> know your quoteing a quote of a quote as it were.
>
> ============================
>
> I know the argument that ...
>
> "the nature of the market changed on 3/1/2000 and therefore
> optimizations should be done after that point."
>
> Now there's a classic statement that I haven't heard the likes of
> since people were talking about the fact that we had a "new
economy"
> in the very late 90's and that we were no longer subject to bear
> markets.
>
> To me the "Nature of the Market" is the same as it's always
been. It
> goes up and it goes down. Without the excesses in one direction
or
> another there would be no value to tools like AB and no reason to
> develop market timing systems as buy & hold would be the rule
forever.
>
> This therefore to me is the reason now and as always to not have a
> testing window that's so short that it can't see anything outside
the
> current micro climate. Lots of people in the late 90's did this
and
> got smashed and now that we've had an almost three year long bear
> market, people are starting to do it again.
>
> Once this starts to pervade the mainstream and/or the talking
heads
> at the financial news networks we'll be absolutely positive the
> bottom is at hand or already behind us.
>
>
> Yahoo! Groups Sponsor
> ADVERTISEMENT
>
>
>
>
> 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 the Yahoo! Terms of
Service.
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/
|