Discrete Fourier Transform
Sort of like a polynomial regression on steroids
The idea for this indicator came from reading several research papers about using the discrete fourier transform (DFT) for time series prediction. Additional inspiration came from Michael Pollock’s YouTube channel because he has been using this technique to day-trade the SPX.
Fourier’s theory states that any curve can be approximated by combining the right number of sine waves with the appropriate amplitude and phase angle. The discrete fourier transform (DFT) isolates the individual sine waves that make up the price curve. Once broken out, it should be easier to see how the dominant waves affect stock price. The theory is that the waves will point to a change in direction—not immediately, but more like a sign on the interstate alerting drivers that their exit will be coming up soon.
The green wave line is the average of individual waves combined with each other. The indicator can be set to combine between 1 and 64 waves. The default is set to the top 6 waves. As more waves are combined, the green wave will get closer to actual stock price. It is important to note that the DFT does not forecast / predict prices. However, it may prove useful as a more advanced form of regression analysis.