Technical Analysis Doesn't Work

Status
Not open for further replies.
Quote from rcanfiel:

<i>In this thought-provoking work, David Aronson tests more than 6,400 technical analysis rules and finds that none of them offer statistically significant returns when applied to trading the S&P 500. This result, presented at the end of his work, is not disappointing to dedicated students of technical analysis who draw from the book not a new trading technique but instead take away a new, and more effective, approach to system development and trading. Those seeking the single best indicator or day trading pattern will be disappointed after reading Evidence-Based Technical Analysis, just as they will be disappointed in their trading until they advance beyond seeking the Holy Grail of Trading. </i>
More...

I've read Aronson's book and found one humugeous flaw in his research. All of his tests involved doing the testing in variable environments thus rendering those test worthless. It would be the same as doing any test in a tainted environment. Whatever the conclusions, they were flawed.

You can argue that point till you are blue in the face but any research scientist will tell you exactly what I have just stated.
 
Quote from rcanfiel:

Well, those in favor of the OP TA description have come up with nothing evidentiary yet in this thread, so not seeing much to fight against.
More...

Plenty of evidence exists for those wishing test it on their own. I agree that saying it exists doesn't make it so . . . just like saying it doesn't is just as worthless.
 
Quote from sirinvestalot:

So I guess that fundamental analysis is the only way to go?:confused:
More...

This exact statement is one of the main reasons the markets move so melodically and methodically. There are those wishing to follow like sheep trying to find the simple path which doesn't exist and those willing to bust butt doing the hard work of searching under rocks to find answers. It is that balance that makes the markets move.
 
All of his tests involved doing the testing in variable environments thus rendering those test worthless.
More...

Are you saying that he did not use the same data set or same lookback period to test each indicator? That would be mind blowingly stupid if that was the case. Of course no conclusions could be made from such a study.

Does he test only single indicator performance or does he test combinations of indicators as well? Indicator performance can be improved by filtering with other indicators. There is a diminishing return if a consensus of multiple indicators is used. Probably best to keep things simple and robust by only using one filter indicator.

I still wouldn't expect a good system to give correct signals much more than about 55-60% of the time however. That's why money management and position sizing need to worked into the mix.
 
I have a friend, who has just finished his doctorate, that submitted his thesis: "The Effective use of Trading Rules and Money Management in Futures Markets"

Everything in there was well thought out and made logical sense and his conclusion was that pivots, candles, point & figure and scaling out were all completely ineffective tools in the search for a completely trustworthy system to make money.

He himself is clever enough to know that because the method used to test such a hypothesis is completely systematic, it's 100% impossible to disprove the value of TA.

To the OP, if it were as simple as finding an autonomous and transparent way to indiscriminately use Technical Analysis and continually make profits, this method would inevitably disappear as it got overused/overexposed. It might be something that originally locked in 20% time after time and, as people realised this, they pre-empted or faded it until that margin disappeared completely. The so-called loss of an edge. This is why, although not all TA trading methods need be complex, there has to be something more to it than knowing how to respond to textbook situations.

To put a point to this tangent, I think it is a little naive to wholeheartedly discount TA on the basis that you can't prove a 1-2-3 indiscrimate system works. As someone pointed out before, whether they realise it or not, people use TA to assess what the collective probablility of a market moving from point A to B is.

Someone who is watching a pattern form, which they see once every week/month/year/ten years unfold is in a far better position to test a hypothesis as to whether his TA indicators are telling him something or not, than the guy who wrote your book. A person watching a combination of between 1 and 10 indicators simultaneously lining up (as they had done successfully in the past), has a deeper set of data to test and a better appreciation as to whether what they're seeing has a message or not. Your bank PIN number is only 4 digits long but you still have to plug them in in the right order. You only get 3 goes to get it right as opposed to the one chance in the market. Those odds are good enough to protect your privacy so how complex do you think this game might be :confused:

In the wrong hands TA is at best useless and, at worst, a financial disaster waiting to happen. You can't blanket test it using a simple hypothesis and some popular interpretations as to how your indicators should be used- if human discretion wasn't a factor, EVERYONE would be plugging in their black box and counting their money whilst down the pub.

I wont bang on any longer, I'm just asking you to be a bit more open minded rather than believing what you read if it supports your stance.

There are millions of people out there that have a driving licence but how many of them know how to race a Nascar? Same same with TA.
 
Quote from rcanfiel:

See reply to Hershey post. This is a belief, not a proven reality. There is little evidence that the "professionals" use pure technical indicators or the other methods as described to any benefit whatsoever.

Stating this remains an unproven. Throwing in the snides also does nothing to provide evidence against the assertion.
More...
As has been stated before in at least one other thread, you can't prove TA doesn't ever work, and nobody with a winning TA system is stupid enough to disclose it to you just to prove you're wrong.

This is a foreseeable impasse.

So what's the point of this, the 637th "TA doesn't work" thread?
 
--------------------------------------------------------------------------------
Quote from jack hershey:


Could you post a set of rigorous evidence that is accepted by other knowledgeable individuals in the investment community that supports this view.

Then we can begin this thread with a hypothesis. So far it is only a limited unsubstantiated opinion.
--------------------------------------------------------------------------------

The OP responds a moment ago:

Fine, the Aronson book can be a portion of the basis. But this statement is true for anyone making an originating post.

http://www.amazon.com/Evidence-Base...84637836&sr=8-1

As well-described by one of the reviewers:

In this thought-provoking work, David Aronson tests more than 6,400 technical analysis rules and finds that none of them offer statistically significant returns when applied to trading the S&P 500. This result, presented at the end of his work, is not disappointing to dedicated students of technical analysis who draw from the book not a new trading technique but instead take away a new, and more effective, approach to system development and trading. Those seeking the single best indicator or day trading pattern will be disappointed after reading Evidence-Based Technical Analysis, just as they will be disappointed in their trading until they advance beyond seeking the Holy Grail of Trading.

My response.

Of the 92 studies that look systematically at easily replicable technical trading rules, Park and Irwin (2004) responded comprehensively showing 58 reported positive excess profits, 10 yielded mixed results and 24 reported losses.

Naturally, people tend to report success. Aronson apparently did not have success in his work which ended with no statistically significant returns. Aronson wound up in the category, above citied, that was least occurring (10 mixed results).

So the tab from authoritative folks at present is:

58 got positive excess profits (this is better profits than the norm)

11 got mixed results (One is Aronson's)

24 reported losses.

So as the OP you are down about 23 authoritative reports to get to indecisive where Aronson resides. To get to where I reside on this hypothesiswise, you have not made one substantive comment so far since Aronson didn't agree with you either.

My response is just a warm up drill. Someone else has already taken the steps to report to you why Aronson failed.

Studies do not represent the meat of the proof that TA works and works well. Also you have placed yourself in a position of your choosing and you may find that it is untenable according to intellectual reasoning. If you do fine, if not, thats okay too.

I have placed myself in the most tenable position for argument and it is also true that I have proof that is well respected by the financial community.
 
Status
Not open for further replies.
Back
Top