Intermittency in Quantitative Finance

TL;DR

This paper applies factorial moments analysis, originally from nuclear physics, to financial time series to detect intermittency and non-Gaussian fluctuations at sub-4-hour resolutions, revealing correlations in price returns.

Contribution

It introduces a novel application of factorial moments to finance, demonstrating their effectiveness in identifying intermittency and correlations in price series at high temporal resolution.

Findings

Intermittent behavior detected below 4-hour resolution.

Factorial moments increase above 1 indicating correlations.

Non-Gaussian fluctuations are sensitive to time scales under 4 hours.

Abstract

Factorial moments are convenient tools in nuclear physics to characterize the multiplicity distributions when phase-space resolution ($\Delta$) becomes small. For uncorrelated particle production within $\Delta$, Gaussian statistics holds and factorial moments $F_q$ are equal to unity for all orders $q$. Correlations between particles lead to a broadening of the multiplicity distribution and to dynamical fluctuations. In this case, the factorial moments increase above 1 with decreasing $\Delta$. This corresponds to what can be called intermittency. In this letter, we show that a similar analysis can be developed on financial price series, with an adequate definition of factorial moments. An intermittent behavior can be extracted using moments of order 2 ($F_2$), illustrating a sensitivity to non-Gaussian fluctuations within time resolution below 4 hours. This confirms that correlations between price returns start to play a role when the time resolution is below this threshold.