On the Informative Content of Technical Indicators: - a Systematic Approach to Pattern Recognition

Abstract: Technical analysis is a rather subjective graphical method of recording the history of trading in a certain stock to deduct the probable future trend in its return. By applying the automatic computer algorithm proposed by Lo, Mamaysky and Wang (2000) on Swedish stock market data we separate ourselves from the subjectivity of ordinary pattern recognition. In this way we are able to systematically determine the informative content of ten commonly used technical indicators over the time-period of 1982 - 2006. At the same time we also perform a large scale out-of-sample test of the ability of the Lo, Mamaysky and Wang (2000) algorithm to detect technical patterns. We conclude that the Lo, Mamaysky and Wang (2000)model performs well, also on Swedish data, and that many of the technical patterns indeed contain incremental information which could be useful for an investment strategy.