Testing the causality of Hawkes processes with time reversal

DOI10.1088/1742-5468/AAAC3FMaRDI QIDQ4964526zbMATH OpenOpenAlexWikidataFDO

Authors Marcus Cordi, Ioane Muni Toke, Damien Challet

Publication date 2 March 2021

Published in Journal of Statistical Mechanics: Theory and Experiment (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1709.08516

Mathematics Subject Classification ID

Point processes (e.g., Poisson, Cox, Hawkes processes) (60G55) Non-Markovian processes: hypothesis testing (62M07) Applications of statistics to actuarial sciences and financial mathematics (62P05)

Abstract: We show that univariate and symmetric multivariate Hawkes processes are only weakly causal: the true log-likelihoods of real and reversed event time vectors are almost equal, thus parameter estimation via maximum likelihood only weakly depends on the direction of the arrow of time. In ideal (synthetic) conditions, tests of goodness of parametric fit unambiguously reject backward event times, which implies that inferring kernels from time-symmetric quantities, such as the autocovariance of the event rate, only rarely produce statistically significant fits. Finally, we find that fitting financial data with many-parameter kernels may yield significant fits for both arrows of time for the same event time vector, sometimes favouring the backward time direction. This goes to show that a significant fit of Hawkes processes to real data with flexible kernels does not imply a definite arrow of time unless one tests it.

Recommendations

• Uncovering causality from multivariate Hawkes integrated cumulants
• Multivariate Hawkes processes: an application to financial data
• Quadratic Hawkes processes for financial prices
• Mean-field inference of Hawkes point processes
• Multivariate quadratic Hawkes processes—part I: theoretical analysis

Cites work

• A GENERALIZED CONTAGION PROCESS WITH AN APPLICATION TO CREDIT RISK
• A Limited Memory Algorithm for Bound Constrained Optimization
• A Simplex Method for Function Minimization
• An estimation procedure for the Hawkes process
• An Introduction to the Theory of Point Processes
• Integrability of Expected Increments of Point Processes and a Related Random Change of Scale
• Locally stationary Hawkes processes
• Market heterogeneities and the causal structure of volatility
• On Lewis' simulation method for point processes
• Point processes and queues. Martingale dynamics
• Quadratic Hawkes processes for financial prices
• Self-exciting point process modeling of crime
• Stochastic Declustering of Space-Time Earthquake Occurrences
• The limits of statistical significance of Hawkes processes fitted to financial data
• Time reversal invariance in finance

Cited in

(4)

• Simultaneous multivariate Hawkes-type point processes and their application to financial markets
• Quadratic Hawkes processes for financial prices
• Uncovering causality from multivariate Hawkes integrated cumulants
• Validity of Time Reversal for Testing Granger Causality

Describes a project that uses

Uses Software

• LBFGS-B

This page was built for publication: Testing the causality of Hawkes processes with time reversal

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4964526)