FractalCycles
Statistically significant market cycles via signal processing
Overview
FractalCycles is a web app that uncovers repeating, statistically significant cycles hidden in market prices. You bring price data for a stock, index, currency or crypto pair, and it breaks that series down into the cycles underneath it, tests each one to see whether it is real, and projects a composite cycle forward on the chart so you can see where the combined rhythm points next. The method is signal processing rather than machine learning, and we are careful not to describe it as AI. It uses the Goertzel discrete Fourier transform to find candidate cycle lengths, the Bartels test to check whether each cycle is statistically significant rather than noise, and the Hurst exponent to judge whether a market is currently trending, mean reverting or behaving randomly. The approach builds on J. M. Hurst's work on market cycles. It is built for self-directed investors, market analysts and researchers who want to understand market structure. It does not give buy or sell signals and it does not promise profits. Every cycle is shown with its significance statistics so the user can judge the evidence rather than take the output on faith. It runs in the browser with nothing to install, and there is a permanent free tier that does not ask for a card.
Details
Similar Tools
- RoboFin Investment analysis
Raising capital?
CrackTheDeck analyzes your pitch deck before investor meetings — structure, narrative, and fit.
Analyze My Deck →