[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

Yeah, well I do have a tendency to ramble a bit when I get going…

 

People taking the message the wrong way is one of the problems with using quotes. For instance, I recently wanted to use a bit of Oscar Wilde myself (on another thread) and quote that “a cynic is the man who knows the price of everything and the value of nothing”…thank God I didn’t do that, eh? Imagine the trouble it would have landed me in!!

 

I agree with most of what you say, especially about the “confidence”, but would just add that a “good trader” (as you put it) is never complacent. A holy grail may not exist, but that doesn’t mean that he/she shouldn’t still look for one! I’m sure it doesn’t matter who you are, your trading strategy can always be improved. That’s why I like Ehler’s stuff – I understand it, I agree with the principles and I think that I might benefit from those “few extra bars” that were talked about. The same goes for adaptive indicators…even if they only test out a “bit” better, then I’m up for it. Providing they don’t over-complicate or compromise the rest of the strategy.

 

Ehler is into finding “theoretically sound principles”. Well thank God someone is!!! I despise reading books authored by people who only have a tentative grasp on the theory behind what they are trying to convey. All too often people/traders mouth off about what “works”…justifying this on nothing but their own experience and a few trading records. Well, OK, so Joe Blogs made this new fancy strategy or indicator work. But WHY did it work? Does he know? Will the success continue? Under what conditions does it work best (or not at all)?

 

I would much rather take a bunch of theory from someone who can explain it well enough so that I understand it…and then worry about the practical implementation of the said theory myself. I want nothing more from Ehler than to come up with a few neat ideas which make theoretical sense. I’m grown up and experienced enough in trading myself (and I’m sure I’m not alone here!) to determine whether or not I can make the theory work in real life. In the meantime, I thank the author for bringing the possibility of a “new way” to my attention. I really do not care if he/she can trade or not.

 

 

 


From: Metastockusers@xxxxxxxxxxxxxxx [mailto:Metastockusers@xxxxxxxxxxxxxxx] On Behalf Of jawjahtek
Sent: Sunday, July 03, 2005 10:51 PM
To: Metastockusers@xxxxxxxxxxxxxxx
Subject: [Metastockusers] Re: Stationarity and Real World application of statistics

 

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