|
PureBytes Links
Trading Reference Links
|
It looks like this thread is touching on equity feedback, a favorite topic
of mine. The idea is to process the equity curve exactly like you would a
price chart, i.e. develop a trading system for it. The trading system would
basically be a trend following systems, oscillator systems do not work on
equities. Some of DT's systems, which have longer periods, may still work on
it. Equity doesn't have OHLC but you can recreate those easily so that you
can use the full spectrum of indicators on the equity.
I attach a simple example chart of what Equity feedback "could do" for you.
Blue is the final equity curve, Black and Red is the stock price, Green is
my feedback indicator. This is a very simple test, not a finished product -
it is intended only to show the equity feedback potential. There are
problems to solve!
AmiBroker allows you to use the equity() as often as you like in your
trading system and will calculate the equity based on the last values of the
buy/sell signals. This means you can easily cascade TWO trading systems:
your Primary system to trade your stock normally and produce the primary
equity curve (you do NOT trade on this system), a Secondary system that is
designed to trade the equity. The Secondary system would reverse Primary
trading signals (now your actual signals) and turn DrawDowns into profit
opportunities.
When using equity feedback "high system performance" and "high robustness"
can actually be a handicap, and de-tuning your system can actually improve
returns. It's a new way of thinking. What you would want is a system the is
extremely sharply tuned to the stock's rhythm, so much so that performance
drops sharply when the stock looses character. The more and the bigger the
up and down waves (DrawDowns) in your equity the better the returns. Anyway,
I could ramble on...this is a big topic.
Take care,
Herman.
-----Original Message-----
From: dtsokakis [mailto:TSOKAKIS@x...]
Sent: Friday, July 26, 2002 9:23 AM
To: amibroker@xxxxxxxxxxxxxxx
Subject: [amibroker] Re:Money management
Anthony,
I thought to begin this thread with a method I use for years, even
with MSEXCEL the ...happy 99 times [I sold the whole thing when the
portfolio equity crossed its 15-day MA, after a long extra-bullish
period, no doubt about it]
Do you critisize the cross EMA40 method or the Equity itself ?
In the second case, we will loose the method [I hope to see other
methods too].
I would like your opinion for this EMA40 idea.
Of course, any other criticism is always appreciated.
The Smoothed Stochastic CCI Equity curve is another [perhaps
interesting] subject.
I will post later the full formula to see better.
DT
--- In amibroker@xxxx, Anthony Faragasso <ajf1111@xxxx> wrote:
> Dimitri,
>
> Thank you for continuing this Thread, It is an important part of any
> traders success or failure.
>
> Strategy and Money management are the two most important parts of
any
> trader's overall plan.
>
> The best entry rule is useless without proper risk control. You can
> almost perfectly analyze a developing market situation, find the
best
> strategy to exploit that situation, and be almost perfectly correct
in
> your forcast of what that market will do, and yet still lose money
if
> you do not use proper risk control and money management.
>
> There are so many variables which constitute Money management , that
> just pinning it to an Equity curve crossover would be dangerous for
most
> traders. In the 3 gifs that you have posted , the Drawdowns in the
> equity curves appear to be excessive even though the equity curves
are
> above the 40 period EMA , how do you protect yourself from these
> drawdowns ? Are you in that much cash to absorb these drawdowns ?
>
>
> Anthony
>
>
>
> Dimitris Tsokakis wrote:
>
> > Another [interesting] example.Athens SE General Index had a nice
> > fitting to the Stochastic CCI system from A [Aug 2000] toB [April
> > 2001]. Take the Profits [nearly +60%] and stay in cash.The 40-day
EMA
> > cross at B is more than clear.The system is no good anymore for a
> > quite long period.Slightest attempts for the Equity red curve to
> > exceed its EMA were very dangerous until mid December 2001.A new
> > fitting period seem to begin and give some interesting profits
[+10%]
> > till Feb 2002 and out of thesystem again and again.The actual
Equity
> > curve should point 17000, the equity without this type of
management
> > is at 7276.The all-season "blind backtesting" has no relation with
> > real trading conditions.Any excellent system may change.You
should be
> > there to stop it, instead of insisting with some fanatism and
loose
> > the profits and a part of theinitial equity in the name of the
holly
> > system.Dimitris Tsokakis
> >
> >
> > Yahoo! Groups Sponsor
> ADVERTISEMENT
>
> >
> > Your use of Yahoo! Groups is subject to the Yahoo! Terms of
Service.
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
------=_NextPart_001_000D_01C2349E.096CD640
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD><TITLE></TITLE>
<META content=3D"text/html; charset=3Diso-8859-1" http-equiv=3DContent-Type=
>
<META content=3D"MSHTML 5.00.2314.1000" name=3DGENERATOR></HEAD>
<BODY>
<P><FONT size=3D2>Hi DT,</FONT></P>
<P><FONT size=3D2>It looks like this thread is touching on equity feedback,=
a=20
favorite topic of mine. The idea is to process the equity curve exactly lik=
e you=20
would a price chart, i.e. develop a trading system for it. The trading syst=
em=20
would basically be a trend following systems, oscillator systems do not wor=
k on=20
equities. Some of DT's systems, which have longer periods, may still work o=
n it.=20
Equity doesn't have OHLC but you can recreate those easily so that you can =
use=20
the full spectrum of indicators on the equity.</FONT></P>
<P><FONT size=3D2>I attach a simple example chart of what Equity feedback "=
could=20
do" for you. Blue is the final equity curve, Black and Red is the stoc=
k=20
price, Green is my feedback indicator. This is a very simple test, not a=20
finished product - it is intended only to show the equity feedback potentia=
l.=20
There are problems to solve! </FONT></P>
<P><FONT size=3D2>AmiBroker allows you to use the equity() as often as you =
like in=20
your trading system and will calculate the equity based on the last values =
of=20
the buy/sell signals. This means you can easily cascade TWO trading systems=
:=20
your Primary system to trade your stock normally and produce the primary eq=
uity=20
curve (you do NOT trade on this system), a Secondary system that is=20
designed to trade the equity. The Secondary system would reverse Prima=
ry=20
trading signals (now your actual signals) and turn DrawDowns into profit=20
opportunities.</FONT></P>
<P><FONT size=3D2>When using equity feedback "high system performance" and =
"high=20
robustness" can actually be a handicap, and de-tuning your system can actua=
lly=20
improve returns. It's a new way of thinking. What you would want is a syste=
m the=20
is extremely sharply tuned to the stock's rhythm, so much so that performan=
ce=20
drops sharply when the stock looses character. The more and the bigger the =
up=20
and down waves (DrawDowns) in your equity the better the returns. Anyw=
ay, I=20
could ramble on...this is a big topic.</FONT></P>
<P><FONT size=3D2>Take care</FONT><FONT size=3D2>,<BR>Herman.</FONT><FONT=20
size=3D2><BR><BR>-----Original Message-----<BR>From: dtsokakis [<A=20
href=3D"mailto:TSOKAKIS@xxxx";>mailto:TSOKAKIS@x...</A>]<BR>Sent: =
Friday,=20
July 26, 2002 9:23 AM<BR>To: amibroker@xxxxxxxxxxxxxxx<BR>Subject: [amibrok=
er]=20
Re:Money management<BR><BR><BR>Anthony,<BR>I thought to begin this thread w=
ith a=20
method I use for years, even<BR>with MSEXCEL the ...happy 99 times [I sold =
the=20
whole thing when the<BR>portfolio equity crossed its 15-day MA, after a lon=
g=20
extra-bullish<BR>period, no doubt about it]<BR>Do you critisize the cross E=
MA40=20
method or the Equity itself ?<BR>In the second case, we will loose the meth=
od [I=20
hope to see other<BR>methods too].<BR>I would like your opinion for this EM=
A40=20
idea.<BR>Of course, any other criticism is always appreciated.<BR>The Smoot=
hed=20
Stochastic CCI Equity curve is another [perhaps<BR>interesting] subject.<BR=
>I=20
will post later the full formula to see better.<BR>DT<BR>--- In amibroker@xxxx=
...,=20
Anthony Faragasso <ajf1111@xxxx> wrote:<BR>> Dimitri,<BR>><BR>&=
gt;=20
Thank you for continuing this Thread, It is an important part of any<BR>>=
;=20
traders success or failure.<BR>><BR>> Strategy and Money management a=
re=20
the two most important parts of<BR>any<BR>> trader's overall=20
plan.<BR>><BR>> The best entry rule is useless without proper risk=20
control. You can<BR>> almost perfectly analyze a developing market=
=20
situation, find the<BR>best<BR>> strategy to exploit that situation, and=
be=20
almost perfectly correct<BR>in<BR>> your forcast of what that market wil=
l do,=20
and yet still lose money<BR>if<BR>> you do not use proper risk control a=
nd=20
money management.<BR>><BR>> There are so many variables which constit=
ute=20
Money management , that<BR>> just pinning it to an Equity curve crossove=
r=20
would be dangerous for<BR>most<BR>> traders. In the 3 gifs that yo=
u=20
have posted , the Drawdowns in the<BR>> equity curves appear to be=
=20
excessive even though the equity curves<BR>are<BR>> above the 40 period =
EMA=20
, how do you protect yourself from these<BR>> drawdowns ? Are you =
in=20
that much cash to absorb these drawdowns ?<BR>><BR>><BR>>=20
Anthony<BR>><BR>><BR>><BR>> Dimitris Tsokakis wrote:<BR>><BR=
>>=20
> Another [interesting] example.Athens SE General Index had a nice<BR>&g=
t;=20
> fitting to the Stochastic CCI system from A [Aug 2000] toB [April<BR>&=
gt;=20
> 2001]. Take the Profits [nearly +60%] and stay in cash.The=20
40-day<BR>EMA<BR>> > cross at B is more than clear.The system is no g=
ood=20
anymore for a<BR>> > quite long period.Slightest attempts for the Equ=
ity=20
red curve to<BR>> > exceed its EMA were very dangerous until mid Dece=
mber=20
2001.A new<BR>> > fitting period seem to begin and give some interest=
ing=20
profits<BR>[+10%]<BR>> > till Feb 2002 and out of thesystem again and=
=20
again.The actual<BR>Equity<BR>> > curve should point 17000, the equit=
y=20
without this type of<BR>management<BR>> > is at 7276.The all-season "=
blind=20
backtesting" has no relation with<BR>> > real trading conditions.Any=
=20
excellent system may change.You<BR>should be<BR>> > there to stop it,=
=20
instead of insisting with some fanatism and<BR>loose<BR>> > the profi=
ts=20
and a part of theinitial equity in the name of the<BR>holly<BR>> >=20
system.Dimitris Tsokakis<BR>> ><BR>> ><BR>>=20
>  =
; =20
Yahoo! Groups=20
Sponsor<BR>> =
&nb=
sp; =20
ADVERTISEMENT<BR>><BR>> ><BR>> > Your use of Yahoo! Groups i=
s=20
subject to the Yahoo! Terms of<BR>Service.<BR><BR><BR>---------------------=
---=20
Yahoo! Groups Sponsor ---------------------~--><BR>Will You Find True=20
Love?<BR>Will You Meet the One?<BR>Free Love Reading by phone!<BR><A=20
href=3D"http://us.click.yahoo.com/7dY7FD/R_ZEAA/Ey.GAA/GHeqlB/TM"=20
target=3D_blank>http://us.click.yahoo.com/7dY7FD/R_ZEAA/Ey.GAA/GHeqlB/TM</A=
><BR>---------------------------------------------------------------------~=
-><BR><BR><BR><BR>Your=20
use of Yahoo! Groups is subject to <A href=3D"http://docs.yahoo.com/info/te=
rms/"=20
target=3D_blank>http://docs.yahoo.com/info/terms/</A><BR><BR><BR></P></FONT=
></BODY></HTML>
------=_NextPart_001_000D_01C2349E.096CD640--
Attachment:
jpg00130.jpg
Attachment:
Description: "Description: JPEG image"
|