Goodness-of-Fit Test for Self-Exciting Processes

Recently there have been many research efforts in developing generative models for self-exciting point processes, partly due to their broad applicability for real-world applications, notably self- and mutual- exciting point processes. However, rarely can we quantify how well the generative model captures the nature or ground-truth since it is usually unknown. The challenge typically lies in the fact that the generative models typically provide, at most, good approximations to the ground-truth (e.g., through the rich representative power of neural networks), but they cannot be precisely the ground-truth. We thus cannot use the classic goodness-of-fit test framework to evaluate their performance. In this paper, we provide goodness-of-fit tests for generative models by leveraging a new connection of this problem with the classical statistical theory of mismatched maximum-likelihood estimator (MLE). We present a non-parametric self-normalizing test statistic for the goodness-of-fit test based on Generalized Score (GS) statistics. We further establish asymptotic properties for MLE of the Quasi-model (Quasi-MLE), asymptotic $\chi^2$ null distribution and power function of GS statistic. Numerical experiments validate the asymptotic null distribution as well as the consistency of our proposed GS test.

[1]  Y. Ogata The asymptotic behaviour of maximum likelihood estimators for stationary point processes , 1978 .

[2]  D. Marsan,et al.  Extending Earthquakes' Reach Through Cascading , 2008, Science.

[3]  S. Wiemer,et al.  Earthquake Likelihood Model Testing , 2007 .

[4]  Jason Eisner,et al.  The Neural Hawkes Process: A Neurally Self-Modulating Multivariate Point Process , 2016, NIPS.

[5]  Y. Ogata Seismicity Analysis through Point-process Modeling: A Review , 1999 .

[6]  Peter Szolovits,et al.  MIMIC-III, a freely accessible critical care database , 2016, Scientific Data.

[7]  Marc Hoffmann,et al.  A recursive point process model for infectious diseases , 2017, Annals of the Institute of Statistical Mathematics.

[8]  G. Székely,et al.  TESTING FOR EQUAL DISTRIBUTIONS IN HIGH DIMENSION , 2004 .

[9]  R. Jennrich Asymptotic Properties of Non-Linear Least Squares Estimators , 1969 .

[10]  Calyampudi R. Rao,et al.  Linear Statistical Inference and Its Applications. , 1975 .

[11]  George E. Tita,et al.  Self-Exciting Point Process Modeling of Crime , 2011 .

[12]  Yosihiko Ogata,et al.  Statistical Models for Earthquake Occurrences and Residual Analysis for Point Processes , 1988 .

[13]  Leonhard Held,et al.  Power-law models for infectious disease spread , 2013, 1308.5115.

[14]  D. Vere-Jones,et al.  Stochastic Declustering of Space-Time Earthquake Occurrences , 2002 .

[15]  Arthur Gretton,et al.  A Test of Relative Similarity For Model Selection in Generative Models , 2015, ICLR.

[16]  Jure Leskovec,et al.  Meme-tracking and the dynamics of the news cycle , 2009, KDD.

[17]  T. Taimre,et al.  Hawkes Processes , 2015, 1507.02822.

[18]  Robert Bartle,et al.  The Elements of Real Analysis , 1977, The Mathematical Gazette.

[19]  A. Veen,et al.  Estimation of Space–Time Branching Process Models in Seismology Using an EM–Type Algorithm , 2006 .

[20]  Frederic Paik Schoenberg,et al.  Facilitated Estimation of ETAS , 2013 .

[21]  A. G. Hawkes,et al.  Performance of information criteria for selection of Hawkes process models of financial data , 2017 .

[22]  R. Engle Wald, likelihood ratio, and Lagrange multiplier tests in econometrics , 1984 .

[23]  Shidong Zhou,et al.  On the Monotonicity, Log-Concavity, and Tight Bounds of the Generalized Marcum and Nuttall $Q$-Functions , 2010, IEEE Transactions on Information Theory.

[24]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[25]  L. Baringhaus,et al.  On a new multivariate two-sample test , 2004 .

[26]  Roger D. Peng,et al.  A Space–Time Conditional Intensity Model for Evaluating a Wildfire Hazard Index , 2005 .

[27]  H. White,et al.  NONLINEAR REGRESSION ON CROSS-SECTION DATA , 1980 .

[28]  F. Schoenberg,et al.  Assessment of Point Process Models for Earthquake Forecasting , 2013, 1312.5934.

[29]  Erik A. Lewis,et al.  Self-exciting point process models of civilian deaths in Iraq , 2011, Security Journal.

[30]  H. White Maximum Likelihood Estimation of Misspecified Models , 1982 .

[31]  Frederic Paik Schoenberg,et al.  Multidimensional Residual Analysis of Point Process Models for Earthquake Occurrences , 2003 .

[32]  Bernhard Schölkopf,et al.  A Kernel Two-Sample Test , 2012, J. Mach. Learn. Res..

[33]  Gentry White,et al.  Self-exciting hurdle models for terrorist activity , 2012, 1203.3680.

[34]  Alex Reinhart,et al.  A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications , 2017, Statistical Science.

[35]  Eric Warren Fox,et al.  Spatially inhomogeneous background rate estimators and uncertainty quantification for nonparametric Hawkes point process models of earthquake occurrences , 2016 .

[36]  Vinayak A. Rao,et al.  A Stein-Papangelou Goodness-of-Fit Test for Point Processes , 2019, AISTATS.

[37]  A. Hawkes Spectra of some self-exciting and mutually exciting point processes , 1971 .

[38]  A. Hawkes,et al.  A cluster process representation of a self-exciting process , 1974, Journal of Applied Probability.

[39]  Zaïd Harchaoui,et al.  Signal Processing , 2013, 2020 27th International Conference on Mixed Design of Integrated Circuits and System (MIXDES).

[40]  Arthur Gretton,et al.  A Kernel Test of Goodness of Fit , 2016, ICML.

[41]  Jiancang Zhuang,et al.  Next-day earthquake forecasts for the Japan region generated by the ETAS model , 2011 .

[42]  D. Boos On Generalized Score Tests , 1992 .

[43]  A. Hawkes Point Spectra of Some Mutually Exciting Point Processes , 1971 .