Market efficiency assumes that at any given time prices fully reflect all available information and the market comprises a large number of rational investors. According to this approach no investor has an advantage in predicting a return on an asset. There are three forms in which the hypothesis is stated: weak; semi-strong; and strong. In various degrees it emerges that no excess returns can be earned using technical analysis, historical prices, or other data.

Speculators will try to exploit anomalies until they disappear. Predictable pattern of price movements eventually will not be traded because transactions costs outweigh benefits. Large and liquid markets where information is widely available should be more efficient. In order to implement investment strategies based on the exploitation of these inefficiencies, transaction costs have to be lower than the expected profits.

Inefficiencies come and go; some may remain for longer periods. Anomalies exist and will continue to exist because investors do not always behave rationally. I believe that in certain markets the main players adopt similar strategies and influence market behavior, leaving niches to be exploited by more flexible and fast traders. This should be a driver when trying to identify new anomalies.





Speak your mind

1 Comment so far

  1. Gary Rogan on August 28, 2007 1:58 pm

    I’ve always thought that the hypothesis is meaningless. Any particular trader doing any sequence of trades will have returns that are generally different than the market. I don’t believe it’s possible to definitively answer if their under or over-performance is due to luck, taking advantage of temporary or permanent anomalies, “poor” judgment, “excessive” fees, or some other form of skill or lack thereof.

    Other than saying that it’s difficult to beat the market, or that some sort of average of all performances, if you disregard the transaction costs, will be equal to the overall market performance, I don’t know what else the theory is saying other than one shouldn’t even try.


Resources & Links