Evidence for Technical Analysis

Question

In finance, technical analysis is a security analysis methodology for forecasting the direction of prices through the study of past market data, primarily price and volume.

Is there any evidence (public available) that technical analysis actually works?

Comments are not for extended discussion; this conversation has been moved to chat. — JohnFx, Nov 24 '15 at 00:06
The best proof you can get is to talk to someone that uses TA and consistently makes money over the long term. — Victor, Nov 24 '15 at 06:06
@Victor 2) Even someone consistently making money must be checked with careful attention (pe: in a bull market even simple strategies such as random buy/sell might consistently give great returns) — Mark Messa, Nov 24 '15 at 14:18

score 1 · Answer 1 · answered Nov 23 '15 at 18:03

To answer your original question: There is proof out there.

Here is a paper from the Federal Reserve Bank of St. Louis that might be worth a read. It has a lot of references to other publications that might help answer your question(s) about TA. You can probably read the whole article then research some of the other ones listed there to come up with a conclusion.

Below are some excerpts:

Abstract: This article introduces the subject of technical analysis in the foreign exchange market, with emphasis on its importance for questions of market efficiency. “Technicians” view their craft, the study of price patterns, as exploiting traders’ psychological regularities. The literature on technical analysis has established that simple technical trading rules on dollar exchange rates provided 15 years of positive, risk-adjusted returns during the 1970s and 80s before those returns were extinguished. More recently, more complex and less studied rules have produced more modest returns for a similar length of time. Conventional explanations that rely on risk adjustment and/or central bank intervention do not plausibly justify the observed excess returns from following simple technical trading rules. Psychological biases, however, could contribute to the profitability of these rules. We view the observed pattern of excess returns to technical trading rules as being consistent with an adaptive markets view of the world.

and

The widespread use of technical analysis in foreign exchange (and other) markets is puzzling because it implies that either traders are irrationally making decisions on useless information or that past prices contain useful information for trading. The latter possibility would contradict the “efficient markets hypothesis,” which holds that no trading strategy should be able to generate unusual profits on publicly available information—such as past prices—except by bearing unusual risk. And the observed level of risk-adjusted profitability measures market (in)efficiency. Therefore much research effort has been directed toward determining whether technical analysis is indeed profitable or not. One of the earliest studies, by Fama and Blume (1966), found no evidence that a particular class of TTRs could earn abnormal profits in the stock market. However, more recent research by Brock, Lakonishok and LeBaron (1992) and Sullivan, Timmermann an d White (1999) has provided contrary evidence. And many studies of the foreign exchange market have found evidence that TTRs can generate persistent profits (Poole 6 (1967), Dooley and Shafer (1984), Sweeney (1986), Levich and Thomas (1993), Neely, Weller and Dittmar (1997), Gençay (1999), Lee, Gleason and Mathur (2001) and Martin (2001)).

I gonna take a while to read this paper. But as soon as possible I'll post my comment. By the way, thanks for the answer. — Mark Messa, Nov 23 '15 at 21:43
@MarkMessa With research papers they usually put the/a conclusion at the end - I skip to that to get a brief summary. Then you can dive into the details after. Hope that helps ;) — Ross, Nov 24 '15 at 13:48
This paper appears to be scientifically flawed, in that it performs a post-hoc analysis of successful trading strategies. That would make it suffer from survivor bias. — MSalters, Nov 24 '15 at 13:53
@MSalters Some interesting conclusions of this paper: 1) [TAs] were able to generate excess returns over a long period during the 1970s and 1980s. 2) [TAs] dominates fundamental analysis at short horizons. It seems to be missing a Scientific Control in order to conclude that the excess return happened due to TA and not despite of TA. By the way, I also agree with the survival bias aforementioned. — Mark Messa, Nov 24 '15 at 15:49
Can there be a survival bias on something that you only want to measure if it survives? :) If a system loses money there is no need to add it into the mix of trying to figure out if systems can make money in my opinion. Maybe I'm not understanding survival bias in this situation. — Ross, Nov 24 '15 at 16:35
If several TA analysis are causing losses, of course they should be added into the mix pool. — Mark Messa, Nov 24 '15 at 17:04
Even if non-TA analysis methods were causing losses in the 70s and 80s, they too should be added to the pool. You can't pick your winners after the fact. The survival bias is due to people that stop using loss-making strategies. — MSalters, Nov 25 '15 at 12:26
There's also an additional concern with studying the effect of TA. To the degree that TA is an objective procedure, TA users will be buying the same shares which causes them to rise, and produce short-term profits. When fundamentals kick in, unrealized profits rapidly disappear again, which in turn may cause a coordinated sell-off by TA users. The resulting decrease in price validates the TA conclusion that the stock should be sold. — MSalters, Nov 25 '15 at 12:30

Evidence for Technical Analysis

1 Answers1