2017/10/06 by Y. Samuel Wang, Mathias Drton, Wang, Y. Samuel +1
Computer Science · #Bayesian Modeling and Causal Inference #Computation (stat.CO) #FOS: Computer and information sciences
paper · pdf · doi:10.48550/arxiv.1710.02588
openalex publication_date 2017/10/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We consider linear structural equation models that are associated with mixed\ngraphs. The structural equations in these models only involve observed\nvariables, but their idiosyncratic error terms are allowed to be correlated and\nnon-Gaussian. We propose empirical likelihood (EL) procedures for inference,\nand suggest several modifications, including a profile likelihood, in order to\nimprove tractability and performance of the resulting methods. Through\nsimulations, we show that when the error distributions are non-Gaussian, the\nuse of EL and the proposed modifications may increase statistical efficiency\nand improve assessment of significance.\n