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

Re: [EquisMetaStock Group] (OT) Entire "statistics" discussion has a major fault.



PureBytes Links

Trading Reference Links

"...They isolate and amplify certain aspects of the
base data in ways which may help us understand price movement 
better..."

They also isolate and amplify certain aspects of the data in such a 
way as to create patterns that aren't really there, which serves to 
confuse our understanding of price movement.  

Ironically, back-testing is what proves that the majority of TA 
doesn't add any value.  The bias of a stock analyst is easy to see 
and guard against, but recognizing and protecting against our own 
bias is tougher.  JMO.    


Luck,

Sebastian
     

--- In equismetastock@xxxxxxxxxxxxxxx, "Andrew Tomlinson" 
<andrew_tomlinson@xxxx> wrote:
> 
> The stock price and the IFT of the RSI are not independent 
variables. The
> indicator is derived from the price. Clearly the RSI cannot "cause" 
a change
> in the price as it is a derivative of the price. This is the same 
for any
> indicator.
> 
> Does this mean that TA is meaningless? I don't see why. All 
indicators are
> filters - no more, no less. They isolate and amplify certain 
aspects of the
> base data in ways which may help us understand price movement 
better.
> 
> There's no magic to causation. We know why stock prices move - at 
any one
> point, a share price will rise if there are more dollars being 
spent on
> buying the stock than selling, and vice versa. End of story. The 
analytical
> problem is that the best measure of that net buying or selling is 
the price
> itself.
> 
> When we backtest data we are certainly looking for persistent 
correlations.
> If we think we understand why they happen, then great - but we may 
be wrong
> much of the time anyway. That shouldn't stop us from trading. The 
whole
> point about TA is that its pragmatic - we go with what we see, not 
what some
> (likely bias) analyst is telling us.
> 
> Andrew
> 
> 
> 
> -----Original Message-----
> From: equismetastock@xxxxxxxxxxxxxxx 
[mailto:equismetastock@xxxxxxxxxxxxxxx]
> On Behalf Of sebastiandanconia
> Sent: Wednesday, September 07, 2005 10:23 AM
> To: equismetastock@xxxxxxxxxxxxxxx
> Subject: [EquisMetaStock Group] (OT) Entire "statistics" discussion 
has a
> major fault.
> 
> There seems to be a widespread confusion between causation and 
correlation.
> 
> If a stock price is in an uptrend and the Inverse Fisher Transform 
of RSI is
> greater than 50 (indicating a "buy" signal, depending on 
parameter), that's
> only a correlation between the IFT of RSI and stock price.  The 
stock isn't
> rising because of the indicator, the indicator is rising because of 
the
> stock price and it's the same story with virtually every TA 
indicator.
> 
> So what is the purpose of back-testing this OR ANY indicator if 
it's not the
> cause of stock price movement?  The only things that are being 
generated are
> meaningless correlations.
> 
> 
> Luck to all,
> 
> Sebastian
> 
> 
> 
> 
> 
> ------------------------ Yahoo! Groups Sponsor --------------------
~--> Put
> more honey in your pocket. (money matters made easy).
> http://us.click.yahoo.com/r7D80C/dlQLAA/cosFAA/BefplB/TM
> --------------------------------------------------------------------
~->
> 
> 
> Yahoo! Groups Links




------------------------ Yahoo! Groups Sponsor --------------------~--> 
Help Sudanese refugees rebuild their lives through GlobalGiving.
http://us.click.yahoo.com/hjNroD/EbOLAA/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/