SIDEBAR: Fourier Analysis
Market data is considered to be a complex wave form composed of cycles with different lengths or frequencies. When you look at a price chart, you can see that there are upward trends with countertrend declines or downward trends with countertrend rallies. Each price movement has a varying number of days between consecutive tops and bottoms; in other words, the price action appears to be the sum of cycles with different cycle lengths.
A fast Fourier transform (FFT) of a complex wave form such as market data decomposes the time series into its underlying frequency components. A Fourier power spectrum is simply a plot in histogram form of the cycle power (y-axis) versus frequency (x-axis).