スパース部分的最小二乗回帰による因果媒介分析.

Causal mediation analysis estimates causal effects by focusing on the mediators between cause and outcome. Multiple causally related mediators are often strongly correlated, making the estimation of causal effects difficult. In addition, recent years have seen a number of mediators compared to a sample size. In this paper, we propose a two-step estimation method based on sparse partial least squares regression and pathway lasso. The proposed method can identify the causal pathways among many candidate causal pathways. The effectiveness of the proposed method is shown by simulation studies and a real data analysis. [ABSTRACT FROM AUTHOR]