Eigenfunction martingale estimators for interacting particle systems and their mean field limit

We study the problem of parameter estimation for large exchangeable interacting particle systems when a sample of discrete observations from a single particle is known. We propose a novel method based on martingale estimating functions constructed by employing the eigenvalues and eigenfunctions of the generator of the mean field limit, linearized around the (unique) invariant measure of the mean field dynamics. We then prove that our estimator is asymptotically unbiased and asymptotically normal when the number of observations and the number of particles tend to infinity, and we provide a rate of convergence towards the exact value of the parameters. Finally, we present several numerical experiments which show the accuracy of our estimator and corroborate our theoretical findings, even in the case the mean field dynamics exhibit more than one steady states. AMS subject classifications. 35Q70, 35Q83, 60J60, 62M15, 65C30.

[1]  Grigorios A. Pavliotis,et al.  Parameter Estimation for the McKean-Vlasov Stochastic Differential Equation , 2021, 2106.13751.

[2]  Grigorios A. Pavliotis,et al.  Mean Field Limits for Interacting Diffusions in a Two-Scale Potential , 2017, J. Nonlinear Sci..

[3]  A. Mukherjea,et al.  Real and Functional Analysis , 1978 .

[4]  Sylvain Faure,et al.  Crowds in Equations - An Introduction to the Microscopic Modeling of Crowds , 2018, Crowds in Equations.

[5]  Josselin Garnier,et al.  Consensus Convergence with Stochastic Effects , 2015, ArXiv.

[6]  H. Qiao,et al.  Parameter Estimation of Path-Dependent McKean-Vlasov Stochastic Differential Equations , 2020, Acta Mathematica Scientia.

[7]  Michael Sørensen,et al.  Estimating equations based on eigenfunctions for a discretely observed diffusion process , 1999 .

[8]  S. Tremaine,et al.  Galactic Dynamics , 2005 .

[9]  G. A. Pavliotis,et al.  Parameter Estimation for Multiscale Diffusions , 2007 .

[10]  Marie-Therese Wolfram,et al.  Parameter Estimation for Macroscopic Pedestrian Dynamics Models from Microscopic Data , 2018, SIAM J. Appl. Math..

[11]  B. Goddard,et al.  Noisy bounded confidence models for opinion dynamics: the effect of boundary conditions on phase transitions , 2020, 2009.03131.

[12]  F. Golse.  On the Dynamics of Large Particle Systems in the Mean Field Limit , 2013, 1301.5494.

[13]  Giacomo Garegnani,et al.  Robust Estimation of Effective Diffusions from Multiscale Data , 2021, ArXiv.

[14]  Julian Tugaut,et al.  Phase transitions of McKean–Vlasov processes in double-wells landscape , 2014 .

[15]  Grigorios A. Pavliotis,et al.  Drift Estimation of Multiscale Diffusions Based on Filtered Data , 2021, Foundations of Computational Mathematics.

[16]  Takashi Suzuki,et al.  Free Energy and Self-Interacting Particles , 2005 .

[17]  Florent Malrieu,et al.  Logarithmic Sobolev Inequalities for Some Nonlinear Pde's , 2001 .

[18]  K. Oelschlager.  A Martingale Approach to the Law of Large Numbers for Weakly Interacting Stochastic Processes , 1984 .

[19]  A. Klar,et al.  Particle methods for multi-group pedestrian flow , 2016, 1607.02326.

[20]  Estimation in Interacting Diffusions: Continuous and Discrete Sampling , 2011 .

[21]  S. Benachour,et al.  Nonlinear self-stabilizing processes – II: Convergence to invariant probability , 1998 .

[22]  A. Sznitman.  Topics in propagation of chaos , 1991 .

[23]  N. Yoshida.  Phase Transition from the Viewpoint of Relaxation Phenomena , 2003 .

[24]  J. Gärtner.  On the McKean‐Vlasov Limit for Interacting Diffusions , 1988 .

[25]  On the Diffusive-Mean Field Limit for Weakly Interacting Diffusions Exhibiting Phase Transitions , 2020, Archive for Rational Mechanics and Analysis.

[26]  Grigorios A. Pavliotis,et al.  Eigenfunction martingale estimating functions and filtered data for drift estimation of discretely observed multiscale diffusions , 2021, ArXiv.

[27]  Franccois Delarue,et al.  Uniform in time weak propagation of chaos on the torus , 2021, 2104.14973.

[28]  R. A. Kasonga.  Maximum likelihood theory for large interacting systems , 1990 .

[29]  T. D. Frank,et al.  Nonlinear Fokker-Planck Equations , 2005 .

[30]  G. A. Pavliotis,et al.  Maximum likelihood drift estimation for multiscale diffusions , 2008, 0806.3248.

[31]  Xinping Xiao,et al.  Maximum likelihood estimation of McKean-Vlasov stochastic differential equation and its application , 2016, Appl. Math. Comput..

[32]  Analyticity of the eigenvalues and eigenfunctions of an ordinary differential operator with respect to a parameter , 1974, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[33]  L. Hansen,et al.  Spectral methods for identifying scalar diffusions , 1998 .

[34]  Michael Sørensen,et al.  A Review of Some Aspects of Asymptotic Likelihood Theory for Stochastic Processes , 1994 .

[35]  Kay Giesecke,et al.  Inference for large financial systems , 2018, Mathematical Finance.

[36]  Vladas Sidoravicius,et al.  Stochastic Processes and Applications , 2007 .

[37]  G. A. Pavliotis,et al.  Long-Time Behaviour and Phase Transitions for the Mckean–Vlasov Equation on the Torus , 2018, Archive for Rational Mechanics and Analysis.

[39]  D. Talay,et al.  Nonlinear self-stabilizing processes – I Existence, invariant probability, propagation of chaos , 1998 .

[40]  D. Dawson.  Critical dynamics and fluctuations for a mean-field model of cooperative behavior , 1983 .

[41]  M. Sørensen,et al.  Martingale estimation functions for discretely observed diffusion processes , 1995 .

[42]  L Ellam,et al.  Stochastic modelling of urban structure , 2018, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.