Observed arrival · 2026-08-25
Correlys asks whether your market relationship survives its window
A financial-analysis tool that tests whether correlations hold across stationary, regime-conditioned, corrected, and out-of-sample data.
Field notes
The proposed workflow separates four failure modes that are often compressed into a single correlation statistic: non-stationarity, regime dependence, uncorrected hypothesis searches, and out-of-sample breakdown. Correlys says the underlying series are public and free, while its paid layer covers conditioning, correction, out-of-sample records, and maintenance. The homepage currently exposes the method and pricing logic, but says the product itself is still being built and has no login yet.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue