|
PureBytes Links
Trading Reference Links
|
Dimitris,
You raise an interesting question. Just how long is a sensible
backtest? Personally, I see no point in backtesting to 1960 let
alone 1930. The computer was hardly invented then, never mind the
internet.
What I do like to see is a period of rising prices and a period of
falling prices. I want to include the most recent data, so that
means taking 2000-2003 as my falling period. The obvious choice for
the rising period is now 1997-2000, the most recent relevant period.
Hence my test period is usually 1/1/1997 to yesterday.
I often think it would be beneficial if we all reported results for
the same bunch of criteria, such as:
Date: 1/1/1997 - present
Initial equity: $10,000
Compounded: Yes (or No?)
Commission: 0.5%
Etc. etc.
Of course, I don't expect my preferences to be selected, but a
discussion about a suitable set might be interesting.
After such a discussion, I would happily share my most interesting
backtests (methods & results). Would anyone else?
Steve
--- In amibroker@xxxxxxxxxxxxxxx, "DIMITRIS TSOKAKIS"
<TSOKAKIS@xxxx> wrote:
> Steve,
> I understand your comments, but there is a more tough question: If
we
> make a 10 or 20-year study and it is consistent to some basic
> principles of the indicator and it is , hopefully, profitable,
shall
> we expect a repetition of the 20-year scenario ?
> The 20-year study may be VERY attractive to force us think in this
> way. Is it correct? I really doubt.
> Anyway, i will temporarily forget these hesitations and give some
> long-term descriptions, just for the dialogue. Give me some more
time
> for the examples.
> DT
> > Dimitris,
> >
> > We have visited this area before. The D-Ratio indicator falls
> apart once
> > taken out of the last few years of bearish conditions. See the
> attached
> > chart where D_Ratio for ^VLIC took us out of the ^NDX in mid
> November 1998
> > at ~1460, and kept us on the sidelines as the ^NDX went to ~4700.
> >
> > I know you don't keep data back before 2000, but you should in
my
> > opinion - unless you are sure that the coming year will be like
> 2000-2002
> > and not like 1997-1999! Even if you do not backtest on 1997-1999
> data, you
> > should be prepared to observe the behaviour of your excellent
> indicators
> > during that period.
> >
> > Steve
> >
> >
> > Send BUG REPORTS to bugs@xxxx
> > Send SUGGESTIONS to suggest@xxxx
> > -----------------------------------------
> > 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.
> >
> >
> > Yahoo! Groups Sponsor
> >
> >
> >
> > Send BUG REPORTS to bugs@xxxx
> > Send SUGGESTIONS to suggest@xxxx
> > -----------------------------------------
> > 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.
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get A Free Psychic Reading! Your Online Answer To Life's Important Questions.
http://us.click.yahoo.com/Lj3uPC/Me7FAA/ySSFAA/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/
|