Spectral analysis of stock market prices

The purpose of this article is to apply spectral analysis to six European Stock markets (Germany, France, Italy, The Netherlands, Belgium and the United Kingdom) and the New York Stock Exchange, over the period 1969–1976. For neither series do the estimates suggest deviations from randomness. However, a simple filter rule shows that substantial profits could have been made by a trader in the six European markets. This demonstrates that for testing market efficiency, spectral analysis is far from the best and the conclusion tends to support the hypothesis of ‘white-noise’ in imperfect markets. Cospectral analysis shows the lead and lag relations between the various stock markets under study.