Identifiability and estimation of recursive max-linear models

Nadine Gissibl, Claudia Klüppelberg, Steffen Lauritzen

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

We address the identifiability and estimation of recursive max-linear structural equation models represented by an edge-weighted directed acyclic graph (DAG). Such models are generally unidentifiable and we identify the whole class of DAG s and edge weights corresponding to a given observational distribution. For estimation, standard likelihood theory cannot be applied because the corresponding families of distributions are not dominated. Given the underlying DAG, we present an estimator for the class of edge weights and show that it can be considered a generalized maximum likelihood estimator. In addition, we develop a simple method for identifying the structure of the DAG. With probability tending to one at an exponential rate with the number of observations, this method correctly identifies the class of DAGs and, similarly, exactly identifies the possible edge weights.

Original languageEnglish
Pages (from-to)188-211
Number of pages24
JournalScandinavian Journal of Statistics
Volume48
Issue number1
DOIs
StatePublished - Mar 2021

Keywords

  • Bayesian network
  • causal inference
  • extreme value theory
  • generalized maximum likelihood estimation
  • graphical model
  • structural equation model

Fingerprint

Dive into the research topics of 'Identifiability and estimation of recursive max-linear models'. Together they form a unique fingerprint.

Cite this