On the pitfalls of Gaussian likelihood scoring for causal discovery

10/20/2022
by   Christoph Schultheiss, et al.
0

We consider likelihood score based methods for causal discovery in structural causal models. In particular, we focus on Gaussian scoring and analyze the effect of model misspecification in terms of non-Gaussian error distribution. We present a surprising negative result for Gaussian likelihood scoring in combination with nonparametric regression methods.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/20/2023

Causal Razors

When performing causal discovery, assumptions have to be made on how the...
research
09/04/2019

Likelihood-Free Overcomplete ICA and Applications in Causal Discovery

Causal discovery witnessed significant progress over the past decades. I...
research
03/15/2012

Causal Conclusions that Flip Repeatedly and Their Justification

Over the past two decades, several consistent procedures have been desig...
research
03/09/2018

Scoring Formulation for Multi-Condition Joint PLDA

The joint PLDA model, is a generalization of PLDA where the nuisance var...
research
01/17/2021

Do In-Person Lectures Help? A Study of a Large Statistics Class

Over 1000 students over the course of four semesters were given the opti...
research
10/07/2014

PAC-Bayesian AUC classification and scoring

We develop a scoring and classification procedure based on the PAC-Bayes...

Please sign up or login with your details

Forgot password? Click here to reset