Technical Analysis Doesn't Work

Status
Not open for further replies.
Quote from taowave:

Surf,as you know,TA comes in all different shapes and sizes.Some are more subjective than others.It is unclear to me where you actually draw the line regarding T.A(no pun intended)...If a trader trades a mean reversion strategy,when its time to put the money on the table is it that vastly different than some standard devaition band trade?? Is he trading technically or "statistically"?

I would have to disagree that Stat Arb is subjective...

From the places I have been,(major investment banks and hedge funds),I would say the only subjective element is "news related"..I will preface that statement with the fact that I was a prop/derivatives trader..


ok, makes sense--are people still trading with stat arb in the traditional sense?

thanks,

surf
 
Quote from marketsurfer:

jack,

your intial premise is wrong. TA patterns are NOT defined. What are the exact parameters of a "head and shoulders" so that it may be tested properly?---as one small example.....

regarding this one subjective pattern-- may i suggest this federal reserve's study "the head and shoulder pattern in US equitites" by Carol Osler.

reality is--- it's the random reinforcement that causes these myths to persist-- per carol osler---i agree with her.

regards,

surf


Lets be adult.

When I say patterns are defined; they are.

The prima facia reason that they aredefined is that I DEFINE THEM mathematically.

You may not have taken the time to do this.

When I read papaers by others, I check to see if they have done the defining required as a beginning premise.

In the financial industry, nothing counts towards the bottom line unless it meets the standard of excellence callede "making money". When that test is met, then the results of making money are there to see by tthose who have colleagual relationships where the "rubber meets the road" is understood.

Patterns are all defined mathematically as a starting point. Doing stats on price movement is an exercise in statisitics and has nothing to do with pattern analysis.

I do not expect you to understand anything that I am posting. Just do not worry about it.
 
I wonder if anyone who has found a t.a. pattern that "works" has ever looked for an occurence of the same pattern where it didn't "work."
 
Quote from no. 22:

I wonder if anyone who has found a t.a. pattern that "works" has ever looked for an occurence of the same pattern where it didn't "work."

ha, funny! being that the patterns ONLY exist in the past, and wouldn't exist if they didn't "work" thats a tall order. for instance, it wouldn't be a head and shoulders top, if it wasn't a top....

best,

surf:D
 
Quote from marketsurfer:

ha, funny! being that the patterns exist in the past, and wouldn't exist if they didn't "work" thats a tall order.

best,

surf:D
I see ... they either work all the time or they're not patterns.
 
Quote from jack hershey:



The prima facia reason that they aredefined is that I DEFINE THEM mathematically.
How about defining your P/L mathematically?

Because so far the extent of your on record trading is a 26% loss in a trading contest

Poor math
 
Quote from marketsurfer:

ok, makes sense--are people still trading with stat arb in the traditional sense?

thanks,

surf

Oh sure..."Market neutral" is still in vogue,though this latest "correction" shed some light on the true risk.

I think you may be correct that the supposed pure stat arb boys may be tilting their baskets by as much as 30%...Just a hunch...
 
Quote from alex.samant:

Rcanfiel: Drop the academical talk. I am making "assertions" but i don't need to prove them...How am i supposed to prove to you something that i have gathered over the years and has become a belief?

Then you are offering opinion and belief. You seem to prefer these over evidence, based on the last exchange. Hard evidence shows little value of tested TA. I will take that over your belief, everytime.

Do you think that just for your sake and all the "haters" sake, someone in their right mind would prove all their assertions or show you their track records? Do you think anyone who is successful wants to prove anything by showing track records? To whom? To posters who only disregard anything that is not theirs?

People who make claims but are unwilling to demonstrate or back them up, even by posting some look ahead "predictions" and showing their methods work, are of little interest to me. I don't want their logic. Gernally, I tend to ignore people who have a feeling, beliefs or other fluffy things.
 
Quote from marketsurfer:

name one investment bank that uses TA,as a primary method of trading---- outside of its marketing department, of course

I worked for an Investment Bank, and TA was NOT the primary method.
 
Status
Not open for further replies.
Back
Top