Non-Stationarity in Financial Time Series and Generic Features

TL;DR

This paper investigates the non-stationary nature of financial markets, revealing generic features in return distributions despite limitations of traditional equilibrium models, and explains these findings with a random matrix model.

Contribution

It uncovers universal features in financial return distributions and introduces a random matrix model to explain non-stationarity effects.

Findings

Empirical return distributions show consistent generic features.

Sample observables depend heavily on the evaluation window.

A random matrix model explains the observed non-stationarity.

Abstract

Financial markets are prominent examples for highly non-stationary systems. Sample averaged observables such as variances and correlation coefficients strongly depend on the time window in which they are evaluated. This implies severe limitations for approaches in the spirit of standard equilibrium statistical mechanics and thermodynamics. Nevertheless, we show that there are similar generic features which we uncover in the empirical return distributions for whole markets. We explain our findings by setting up a random matrix model.