Technical analysts exhibit no skill in predicting.....

Discussions about personal psychology for the individual trader.
Post Reply
redbullpeter
Roundtable Fellow
Roundtable Fellow
Posts: 82
Joined: Mon Apr 28, 2003 9:12 am
Location: London, UK

Technical analysts exhibit no skill in predicting.....

Post by redbullpeter » Mon Oct 13, 2003 8:45 am

An article in the FT Weekend, where a researcher tested the ability of respondents to predict future price movements based on a given price series.

http://search.ft.com/search/article.htm ... 1011001313

Interesting but I don't think the conclusions can really be applied to mechanical traders even though we would be classified as technical analysts.

Peter

Kiwi
Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi » Mon Oct 13, 2003 6:47 pm

The articles methodology was not all that relevant in my opinion, whether you are a system trader or discretionary.

In discretionary trading I dont predict the future. All I do is react to a price series and when my setup occurs I enter. I then take the position off according to a set of rules that moves the odds in my favour. If I change either then the expectancy of that particular setup will change. Not really different to a computer system.

Even Elliott Wavers dont really predict the future. Most operate at a couple of points in the wave that are trend following choices or double tops and then take the position off at extensions that are very likely fib relationships with the prior thrust. In most cases this again just amounts to a high probability setup.

My only predictions would be that over the next twenty trades my win ratio should average x, say 70%, and my win loss should be around 2:1. Who knows how the next trade will turn out - or what direction most price series are most likely to take. And for a trader - who cares.

John

bloom
Senior Member
Senior Member
Posts: 35
Joined: Thu Apr 17, 2003 12:45 am
Location: SARS infested HONG KONG..ahh

Post by bloom » Tue Oct 14, 2003 6:30 am

I read the article to the point where it says there is only 19 respondents.
and then I thought I wouldn't want to waste my time anymore..

and it was suppose to be conducted by a statistican..

I didn't bother to read the bottom, tell me if I am miss something..

Jwebster
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Fri Jul 04, 2003 10:23 am
Location: London, UK

FT article

Post by Jwebster » Tue Oct 14, 2003 8:11 am

I agree with Kiwi. The article is potentially useful for showing you there are a lot of windbags out there, but to my mind systems traders shouldn't be concerned about predicting the future accurately, only how they will react to what the market throws at them. The article didn't focus on that aspect, but then if it did it wouldn't have given the media anything to hype.

William
Roundtable Knight
Roundtable Knight
Posts: 238
Joined: Sun May 04, 2003 4:41 pm
Location: Manhattan, New York

Post by William » Tue Oct 14, 2003 12:49 pm

Kiwi,

I couldn't agree more, i started to write out some points of contention about the "field research" and thought it wasn't worth it. I think if you understand how to use technical the right way, prediction isn't the issue. Most people expect to recieve ongoing technical commentary, which most people assume technicals supply...a.k.a - is this stock going up for the next five minutes.

In the end, most every hedge fund that i have visited have charts on their computer screens, i have talked to too many people who swear by it and depend on it to make a living. So, to each his own. Whatever works for you.

Our technical research dept (of which i am a member) resides within a firm that has a strong fundamental bias. Our group has built a following which is larger than the firm's us and global fundamental strategists', in part because of our ability to help them navigate the past 3 years successfully. We didnt know in Oct 2002, that we would be here, but we did realize it was a tradeable bottom and since then the markets have never broken our stops so we've ridden it up. Where it will end, is anybody's guess, but we'll know it when we see it.

MCT
Roundtable Knight
Roundtable Knight
Posts: 102
Joined: Fri May 16, 2003 7:27 pm

Post by MCT » Tue Oct 14, 2003 9:51 pm

John,

I’ll second that. That was an excellent point. As to Elliot Waves I would also add the perspective it presents is of immense value. For instance, it suggests that the market goes three steps forward for every two steps back. If you can manage your fear during the two steps back and your greed during the three steps forward, you’re ahead of the crowd. Such perspective makes it easier to stick to a plan based on reasonable probabilities without allowing either greed of fear to interfere. As you pointed out prediction has nothing to do with it. And of course those few that are competent at reading the patterns, and positioned to catch the trend change, will make more in one year than most will make in five.

MT

AFJ Garner
Roundtable Knight
Roundtable Knight
Posts: 2040
Joined: Fri Apr 25, 2003 3:33 pm
Location: London
Contact:

Philip Coggan

Post by AFJ Garner » Thu Oct 16, 2003 6:22 am

I too read the article; when it originally came out. Mr Coggan is a thoughtful and interesting writer and I have enjoyed his articles over the years. Alpesh Patel can be interesting also - but can also often be wrong and / or misleading. Remember these people HAVE to find things to write about all the time and can not be expected to get it right all the time.

It is the case with the FT however (and indeed with most of the general media) that articles are written by generalists with the intention of entertaining the public at large.

I have found over and over again that such articles are always too simplistic for those of us who have made a deeper study of the topic in question. I found the article perhaps somewhat shallow and partial......but there again a journalist does not attempt to set out a definitive textbook opinion.

For your amusement I should relate that a relative by marriage is the editor of a certain week end section of one of the better known and more serious UK newspapers. She often writes of weekend visits to restaurants, hotels or country pubs. And often she and her husband will have visited such establishements and enjoyed their hospitality, courtesy of an expense account.

What will amuse you however is that she and other journalists just as often write pieces about visits, lunches or holidays which they have never had. It is quite a common practice!

Jim Wells
Contributor
Contributor
Posts: 3
Joined: Sat May 17, 2003 5:20 pm

"field study"

Post by Jim Wells » Sat Oct 18, 2003 2:03 pm

I saw the study (did not participate) and wrote Mr. Burns with some suggestions for a contest and he was kind enough to reply within a few hours. Below is his email to me with most of mine quoted inline.

---

Thanks for your email. You are right that designing
a study for this area is exceedingly hard, most especially
because of the low success rate that everyone will have.

I think that round two should be a lot more acceptable
to a lot more people, but it still needs more work on
the details, and the infrastructure needs to be built.

Regards,

Patrick

James Wells wrote:

>Mr. Burns,
>
>I enjoyed reading your report on the technical analysis study. This is
>clearly an area which merits the very sort of research you are doing. I
>would like to raise two issues which I hope might be useful for future
>research in this area.
>
>Firstly, your method of gathering data from participants seems prone to
>making any method of analysis appear less effective than it actually is.
>Consider, for example, a study on the effectiveness of fundamental
>analysis in which participants are furnished with balance sheets of NYSE
>and NASDAQ listed companies. To test their analysis skill level, each
>participant must select the subsequent price series from a group of four
>choices. I doubt that any sort of analysis, fundamental, technical or
>otherwise would do very well.
>
>Secondly, and this is the reason why I did not participate in the study,
>the decision called for in your study was essentially a test for future
>price prediction accuracy. This, unfortunately, leaves untestable the
>methods of technical analysis which are intended to generate profits from
>having mostly small unprofitable trades and a few large profitable trades.
>(see next para.) The test which would be most effective in evaluating
>these and any other methods would be to track simulated accounts of
>traders with an evaluation of the outcomes of their decisions to buy and
>sell.

Post Reply