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

[Metastockusers] Re: Stationarity and Real World application of statistics



PureBytes Links

Trading Reference Links

teclogeo,

My apologies if the Mark Twain quote came across as a complete 
dismissal of statistics or statisticians. I had meant the comment in 
the same way as I believe Twain wrote it (with humor, irony, and a 
recognition that it will sometimes be true). And I am one of the real 
people that uses statistics in his every day work.

I agree that the whole discussion of Stationarity has gone to far 
(and I was the nut case that started it, shame on me). The point I 
was trying to make is that academically it can be shown that trading 
models that rely on filters (moving averages, transforms, etc.) are 
not the holy grail of trading and that academically the holy grail 
cannot exist. Since this applies to trading with adaptive indicators 
offered here, I thought it was appropriate for this message board.

However, superfragilist put it best:

"Adaptive tools, in general, do test out a bit better than non-
adaptive tools. However, in reality I'm not sure they're going to 
beat a couple of moving averages in live trading. I use several 
adaptive tools and feel comfortable with them and the concepts behind 
them."

If you can confidently trade with adaptive indicators and you believe 
that you will get slightly better performance, then you will be a 
better trader. Confidence in the execution of your trades is far more 
important than any improvement you may (or may not) get from using an 
adaptive indicator or a standard (classic) indicator.

Again, this is the difference between developing indicators and 
trading. I am simply stating my belief that a trading holy grail not 
only doesn't exist, it can't exist. This does NOT affect my trading. 
A good trader accepts indicators and systems that are "good enough". 
A good trader accepts that some things may have to be assumed as 
irrelavent to the trade (then hope like hell that the assumption 
doesn't come back and bite the trader in the pocket book).


jawjahtek


PS: John Ehlers would not agree with my thoughts. He describes his 
job as "finding theoretically sound principles" that can be 
implemented in trading. If it is not theoretically sound, he 
considers it a failure. Classic example of a developer that will 
never be a good trader.



--- In Metastockusers@xxxxxxxxxxxxxxx, "teclogeo" <teclogeo@xxxx> 
wrote:
> Brad,
> 
>  
> 
> I think you're right about the stationarity stuff. Anyway, I didn't 
want to
> hold up geostatistics as particularly having any direct relevance to
> trading. It has its detractors in any case, who hold it to be the 
greatest
> work of evil since Mein Kampf (miners can also be rather fanatical
> sometimes!). I just wanted to point out that statistics is not just
> practiced by politicians and beardy-weirdy university types. Real 
people can
> use it too.!
> 
>  





 
Yahoo! Groups Links

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

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

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