Investigating the efficient markethypothesis using Fourier analysis

Abstract: This study examines if the Swedish stock market adheres to the weak form efficient market hypothesis using Fourier analysis on past stock prices to identify possible cyclic returns. Fourier analysis is well suited for finding seasonalities which would violate the weak-form efficiency. 10 firms were randomly selected from stock market index OMX30 to represent the Swedish stock market. All firms investigated showed signs of periodic behaviour in the long term but adhered to the weak form efficiency in the short term. Some of the cycles found supports some already known calendar effects, such as the U.S presidential election cycle, January effect and ''Sell in May'' strategy. However, the transforms were of poor resolution due to short data sets, making it difficult to differentiate potential cycles from white noise. In addition, this paper does not account for risk and assumes that all price mechanism are accurate.