A multiscale view on inverse statistics and gain/loss asymmetry in financial time series

TL;DR

This paper investigates the gain/loss asymmetry in financial time series across multiple scales, revealing it is a long-term phenomenon linked to correlated downward movements, and introduces a new explanatory model.

Contribution

It applies multiscale analysis using wavelet transforms to uncover the long-term nature of asymmetry and proposes a novel model based on correlated downtrends in stocks.

Findings

Asymmetry disappears when low frequency content is removed.

The asymmetry is a long-term scale phenomenon.

A new model explains asymmetry through prolonged, correlated declines.

Abstract

Researchers have studied the first passage time of financial time series and observed that the smallest time interval needed for a stock index to move a given distance is typically shorter for negative than for positive price movements. The same is not observed for the index constituents, the individual stocks. We use the discrete wavelet transform to illustrate that this is a long rather than short time scale phenomenon -- if enough low frequency content of the price process is removed, the asymmetry disappears. We also propose a new model, which explain the asymmetry by prolonged, correlated down movements of individual stocks.