The Epps effect revisited

We present a simple microstructure model of financial returns that combines (i) the well-known ARFIMA process applied to tick-by-tick returns, (ii) the bid-ask bounce effect, (iii) the fat tail structure of the distribution of returns and (iv) the non-Poissonian statistics of intertrade intervals. This model allows us to explain both qualitatively and quantitatively important stylized facts observed in the statistics of microstructure returns, including the short-ranged correlation of returns, the long-ranged correlations of absolute returns, the microstructure noise and Epps effects. According to the microstructure noise effect, volatility is a decreasing function of the time scale used to estimate it. Paradoxically, the Epps effect states that cross correlations between asset returns are increasing functions of the time scale at which the returns are estimated. The microstructure noise is explained as the result of the negative return correlations inherent in the definition of the bid-ask bounce component (ii). In the presence of a genuine correlation between the returns of two assets, the Epps effect is due to an average statistical overlap of the momentum of the returns of the two assets defined over a finite time scale in the presence of the long memory process (i).

show abstract

“…The microstructure return model introduced in the present paper is able to identify the roots of the Epps effect [11,21], as we now demonstrate.…”

Section: Epps Effect: Basic Notionsmentioning

confidence: 75%

A Simple Microstructure Return Model Explaining Microstructure Noise and Epps Effects

Sornette

Саичев

2012

SSRN Journal

View full text Add to dashboard Cite

show abstract

“…We denote by Σ ij τ the return covariance of contracts i and j at scale τ , defined as Σ we only observe a weak decrease of the variance at short lags in the signature plot, and the ratio between covariance and variance -that determines the so-called Epps effect [18,19] -is almost flat in τ . This is consistent with the absence of statistical arbitrage price, because the time scale for these arbitrage effects is nowadays expected to be well below the five-minute time scale [20][21][22].…”

Section: The Correlation Structure Of Returnsmentioning

confidence: 99%

Dissecting cross-impact on stock markets: an empirical analysis

et al. 2017

View full text Add to dashboard Cite

Abstract. The vast majority of market impact studies assess each product individually, and the interactions between the different order flows are disregarded. This strong approximation may lead to an underestimation of trading costs and possible contagion effects. Transactions in fact mediate a significant part of the correlation between different instruments. In turn, liquidity shares the sectorial structure of market correlations, which can be encoded as a set of eigenvalues and eigenvectors. We introduce a multivariate linear propagator model that successfully describes such a structure, and accounts for a significant fraction of the covariance of stock returns. We dissect the various dynamical mechanisms that contribute to the joint dynamics of assets. We also define two simplified models with substantially less parameters in order to reduce overfitting, and show that they have superior out-of-sample performance.

show abstract

“…• Information needs a human time scale to be processed [2]. On financial markets, some assets, called the leaders, which are often the most liquid, incorporate information onto their prices faster than others, called the laggers.…”

Section: Empirical Fact: the Epps Effectmentioning

confidence: 99%

“…• at high frequency: reproduce the Epps effect [1], take into account lead-lag relationships between assets [2] • at the daily scale: avoid purely Gaussian correlations [3] We develop a theoretical framework based on correlated point processes in order to capture the Epps effect in section 1. We show in section 2 that this model converges to correlated Brownian motions when moving to large time scales.…”

mentioning

confidence: 99%