Persistence and Mean Reversion in Market Data
(Mar 2013)
Jason Voss at the CFA Institute has recently written a very interesting series of posts on the Hurst exponent, which is "a method for detecting persistence, randomness, or mean reversion in financial markets." The Hurst exponent measures the degree to which a signal depends on previous values--a phenomenon known as autocorrelation--and specifically whether values tend to 'switch' (e.g., high values followed by low values) or 'persist' (e.g., high values followed by other high values). Jason...
