Recursive max-linear models with propagating noise

Recursive max-linear vectors model causal dependence between node variables by a structural equation model, expressing each node variable as a max-linear function of its parental nodes in a directed acyclic graph (DAG) and some exogenous innovation. For such a model, there exists a unique minimum DAG, represented by the Kleene star matrix of its edge weight matrix, which identifies the model and can be estimated. For a more realistic statistical modeling we introduce some random observational noise. A probabilistic analysis of this new noisy model reveals that the unique minimum DAG representing the distribution of the non-noisy model remains unchanged and identifiable. Moreover, the distribution of the minimum ratio estimators of the model parameters at their left limits are completely determined by the distribution of the noise variables up to a positive constant. Under a regular variation condition on the noise variables we prove that the estimated Kleene star matrix converges to a matrix of independent Weibull entries after proper centering and scaling.

[1]  P. Butkovic.  Max-linear Systems: Theory and Algorithms , 2010 .

[2]  Andreas Ritter,et al.  Structural Equations With Latent Variables , 2016 .

[3]  S. Lauritzen,et al.  Bayesian Networks for Max-Linear Models , 2019, Network Science.

[4]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[5]  P. Hall,et al.  Moving-maximum models for extrema of time series , 2002 .

[6]  Bernhard Schölkopf,et al.  Causal discovery with continuous additive noise models , 2013, J. Mach. Learn. Res..

[7]  J. Corcoran.  Modelling Extremal Events for Insurance and Finance , 2002 .

[8]  Johan Segers,et al.  A continuous updating weighted least squares estimator of tail dependence in high dimensions , 2016, Extremes.

[9]  Richard L. Smith.  Maximum likelihood estimation in a class of nonregular cases , 1985 .

[10]  Giuseppe F. Italiano,et al.  Amortized Efficiency of a Path Retrieval Data Structure , 1986, Theor. Comput. Sci..

[11]  M. Meerschaert.  Regular Variation in R k , 1988 .

[12]  Sebastian Engelke,et al.  Graphical models for extremes , 2018, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[13]  M. Jirak,et al.  Adaptive function estimation in nonparametric regression with one-sided errors , 2013, 1305.6430.

[14]  Richard A. Davis,et al.  Estimation for first-order autoregressive processes with positive or bounded innovations , 1989 .

[15]  P. Hall,et al.  Nonparametric 'regression' when errors are positioned at end-points , 2009, 0909.0823.

[16]  C. Kluppelberg,et al.  Max-linear models on directed acyclic graphs , 2015, Bernoulli.

[17]  Geert Jan Olsder,et al.  Synchronization and Linearity: An Algebra for Discrete Event Systems , 1994 .

[18]  Steffen L. Lauritzen,et al.  Causal Inference from Graphical Models , 2001 .

[19]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[20]  On causal discovery with an equal-variance assumption , 2018, Biometrika.

[21]  S. Lauritzen,et al.  Identifiability and estimation of recursive max‐linear models , 2019, Scandinavian Journal of Statistics.

[22]  Sidney I. Resnick,et al.  Basic properties and prediction of max-ARMA processes , 1989, Advances in Applied Probability.

[23]  Claudia Klüppelberg,et al.  Estimating an extreme Bayesian network via scalings , 2019, J. Multivar. Anal..

[24]  Steffen L. Lauritzen,et al.  Independence properties of directed markov fields , 1990, Networks.

[25]  Nicole A. Lazar,et al.  Statistics of Extremes: Theory and Applications , 2005, Technometrics.

[26]  Tom Burr,et al.  Causation, Prediction, and Search , 2003, Technometrics.

[27]  Franz von Kutschera,et al.  Causation , 1993, J. Philos. Log..

[28]  J. Kruskal.  On the shortest spanning subtree of a graph and the traveling salesman problem , 1956 .

[29]  Michael I. Jordan.  Graphical Models , 2003 .

[30]  J. Kiefer,et al.  CONSISTENCY OF THE MAXIMUM LIKELIHOOD ESTIMATOR IN THE PRESENCE OF INFINITELY MANY INCIDENTAL PARAMETERS , 1956 .