[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [EquisMetaStock Group] System building - philosophical question



PureBytes Links

Trading Reference Links

Purely by coincidence, one of the money managers I follow addressed 
this issue just today.
   

June 6, 2005 
Recession? Risks Developing, but Not Just Yet 

John P. Hussman, Ph.D.
All rights reserved and actively enforced. 

With job growth of just 78,000, Friday's employment report was 
disappointing, but economists were right to point out that one report 
doesn't make a trend.

Of course, the "one report doesn't make a trend" defense assumes that 
all your information is coming from just one variable (in this case, 
the monthly employment report). If you only use one variable, and it 
has a certain amount of month-to-month variability or "noise," you 
probably are going to need a whole series of reports (a "time 
series") to isolate the true information from the random 
fluctuations....
 
If your information comes from noisy variables like employment data, 
interest rates, or stock prices, you've got a choice: a) sit in a 
lotus position repeating "one report doesn't make a trend" again and 
again, withholding judgment until a trend is obvious even to a blind 
monkey, or b) focus on cross-sectional evidence (movements in many 
variables at the same point in time) rather than just time-series 
evidence (movements in the same variable at many points in time). For 
instance, if you look at whole collection of very good measures of 
economic conditions, and they start simultaneously deteriorating or 
diverging in important ways, you don't need to wait for a whole 
series of reports – you've already got enough information to isolate 
the "signal" from the "noise."

As an example, a lot of mutual fund investors pay attention to the 
200-day moving average of the S&P 500, defining the market to be in 
an uptrend if the S&P is above the moving average, and a downtrend if 
it is below. In effect, investors are filtering out the short-
term "noise" by taking the average of 200 trading days. The problem 
is that following a time series of observations on a single variable 
is an awfully inefficient way of getting your information. By the 
time a new trend is obvious to you, you're probably not alone....
 
http://www.hussmanfunds.com/wmc/wmc050606.htm

It's simply not possible to "improve" one signal from one variable 
using nothing but the data from that variable, and it doesn't matter 
how complex or "sophisticated" the method you use.  The only way to 
to improve the quality of a signal is with an independent 
confirmation from a different variable.  On the bright side, though, 
using multiple variables that confirm each other can give you more 
timely information than by waiting for a time series.


Luck,

Sebastian  






--- In equismetastock@xxxxxxxxxxxxxxx, "Paul Lerno" <paul.lerno@xxxx> 
wrote:
> MG,
> I did an exercise on 294 stocks of Euronext using the roc10
> of MACD for stocks whos's macd was rising over zero.
>  No correlation (-0.07)  is found between the signal and the gains 
10 d later.
> Greetings,
> Paul
>   ----- Original Message ----- 
>   From: mgf_za_1999 
>   To: equismetastock@xxxxxxxxxxxxxxx 
>   Sent: Saturday, June 04, 2005 8:47 PM
>   Subject: Re: [EquisMetaStock Group] System building - 
philosophical question
> 
> 
>   Yes, I am sort of trying to dodge this as it introduces one more
>   variable and one less degree of freedom - the level (2,1 or 0.5) 
of
>   strength.  The nice thing about this approach is that you have a 
range
>   where you will be 'out', say between -1 and +1, and not 
always 'in'
>   the market.  Thanks for the reply!  It sounds as if you have done 
this
>   already.  What did you use for strength?
> 
>   Regards
>   MG Ferreira
>   TsaTsa EOD Programmer and trading model builder
>   http://www.ferra4models.com
>   http://fun.ferra4models.com 
> 
> 
> 
> 
>   --- In equismetastock@xxxxxxxxxxxxxxx, "Paul Lerno" 
<paul.lerno@xxxx>
>   wrote:
>   > Hi MG,
>   > If you have a formula for strengt (?) then you van examine on a
>   great sample what value of strengt is statistical
>   > significant. For instance the 2%,1%,0.5% highest strengt.
>   > I can examine this for the Euronext equities, but its necessary 
to
>   have a mesure for strengt.
>   > Succes.
>   > Paul Lerno (paul.lerno@xxxx)
>   >   ----- Original Message ----- 
>   >   From: mgf_za_1999 
>   >   To: equismetastock@xxxxxxxxxxxxxxx 
>   >   Sent: Saturday, June 04, 2005 11:53 AM
>   >   Subject: [EquisMetaStock Group] System building - 
philosophical
>   question
>   > 
>   > 
>   >   Hi there,
>   > 
>   >   This is a general, almost philosophical question, but some of 
you may
>   >   have had to deal with it before or may have some ideas on 
it.  It
>   >   pertains to automated trading and system building.  If you 
build an
>   >   automated trading system, you often create some 'signal' and 
trade
>   >   based on this.  Now, this signal could be say the difference 
between
>   >   two moving averages, and you trade whenever they cross which 
will be
>   >   exactly when the signal crosses the zero line.
>   > 
>   >   When you evaluate such a system, you buy when the signal goes 
positive
>   >   and sell when it goes negative.  All of this is fairly 
general and
>   >   fairly common.  Note in this example, however, that the 
actual value
>   >   or magnitude of the signal does not play any role - the 
important
>   >   thing is when it crosses the zero line.  Again, no rocket 
science
>   in this.
>   > 
>   >   I am toying with the idea to build a system based on the 
*strength* of
>   >   the signal rather than when it crosses zero, which introduces 
some
>   >   problems.  The question really is, when do you then enter a 
trade?
>   > 
>   >   One obvious way is to put some band around zero and trade 
when the
>   >   signal goes outside of this band.  This complicates matters 
as you
>   >   also have to create this band.  Another way is to still trade 
on the
>   >   zero-line crossover, which is fine but then you could just as 
well
>   >   stick with the original where the whole system is built 
around this. 
>   >   You could use a moving average of the signal line and trade 
when they
>   >   cross, but this is just the same as trading a zero line 
crossover.
>   > 
>   >   What ideas do you have for automatically trading a system 
using some
>   >   signal, that are not triggered by this signal crossing zero?
>   > 
>   >   Regards
>   >   MG Ferreira
>   >   TsaTsa EOD Programmer and trading model builder
>   >   http://www.ferra4models.com
>   >   http://fun.ferra4models.com 
>   > 
>   > 
>   > 
>   > 
>   > 
>   > 
>   > 
>   >
>   ------------------------------------------------------------------
------------
>   >   Yahoo! Groups Links
>   > 
>   >     a.. To visit your group on the web, go to:
>   >     http://groups.yahoo.com/group/equismetastock/
>   >       
>   >     b.. To unsubscribe from this group, send an email to:
>   >     equismetastock-unsubscribe@xxxxxxxxxxxxxxx
>   >       
>   >     c.. Your use of Yahoo! Groups is subject to the Yahoo! 
Terms of
>   Service.
> 
> 
> 
> 
> --------------------------------------------------------------------
----------
>   Yahoo! Groups Links
> 
>     a.. To visit your group on the web, go to:
>     http://groups.yahoo.com/group/equismetastock/
>       
>     b.. To unsubscribe from this group, send an email to:
>     equismetastock-unsubscribe@xxxxxxxxxxxxxxx
>       
>     c.. Your use of Yahoo! Groups is subject to the Yahoo! Terms of 
Service.




------------------------ 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/BefplB/TM
--------------------------------------------------------------------~-> 

 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/equismetastock/

<*> To unsubscribe from this group, send an email to:
    equismetastock-unsubscribe@xxxxxxxxxxxxxxx

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/