Reasoning of Causal Direction in Linear Model Based on Spearman's Rank Correlation Coefficient

Currently, the mining of causality has drawn enormous attention in artificial intelligence. The paper mainly focuses on the causal direction inference problem from an observational sample of the joint distribution in a linear model where the data contain less asymmetric information compared to nonlinear situation. The paper studies the linear additive noise model and analyses the inferring conditions for linear causal direction inference. This paper proposes the copula for modeling dependence and presents a new causal inference method based on Spearman's rank correlation coefficient. The performance of the proposed method is verified through the experiments and analysis on both simulated data and real-world data.