Computing maximum likelihood estimates in recursive linear models with correlated errors

Abstract

In recursive linear models, the multivariate normal joint distribution of all variables exhibits a dependence structure induced by a recursive (or acyclic) system of linear structural equations. These linear models have a long tradition and appear in seemingly unrelated regressions, structural equation modelling, and approaches to causal inference. They are also related to Gaussian graphical models via a classical representation known as a path diagram. Despite the models' long history, a number of problems remain open. In this paper, we address the problem of computing maximum likelihood estimates in the subclass of `bow-free' recursive linear models. The term `bow-free' refers to the condition that the errors for variables $i$ and $j$ be uncorrelated if variable $i$ occurs in the structural equation for variable $j$. We introduce a new algorithm, termed Residual Iterative Conditional Fitting (RICF), that can be implemented using only least squares computations. In contrast to existing algorithms, RICF has clear convergence properties and finds parameter estimates in closed form whenever possible.

22 pages; removed an incorrect identifiability claim

References in corpus (2)

• Likelihood ratio tests and singularities
• Markov equivalence for ancestral graphs

Cited by in corpus (12)

• Global identifiability of linear structural equation models
• Differentiable Causal Discovery Under Unmeasured Confounding
• A factorization criterion for acyclic directed mixed graphs
• Bayesian network structure learning with causal effects in the presence of latent variables
• Algebraic Equivalence of Linear Structural Equation Models
• Integer Programming for Causal Structure Learning in the Presence of Latent Variables
• Efficient least squares for estimating total effects under linearity and causal sufficiency
• Stability of Linear Structural Equation Models of Causal Inference
• Introducing Gaussian covariance graph models in genome-wide prediction
• Deconfounded Score Method: Scoring DAGs with Dense Unobserved Confounding
• Path Dependent Structural Equation Models
• Robust Identifiability in Linear Structural Equation Models of Causal Inference