High-Dimensional Mediation Analysis Based on Additive Hazards Model for Survival Data

Mediation analysis has been extensively used to identify potential pathways between exposure and outcome. However, the analytical methods of high-dimensional mediation analysis for survival data are still yet to be promoted, especially for non-Cox model approaches. We propose a procedure including “two-step” variable selection and indirect effect estimation for the additive hazards model with high-dimensional mediators. We first apply sure independence screening and smoothly clipped absolute deviation regularization to select mediators. Then we use the Sobel test and the BH method for indirect effect hypothesis testing. Simulation results demonstrate its good performance with a higher true-positive rate and accuracy, as well as a lower false-positive rate. We apply the proposed procedure to analyze DNA methylation markers mediating smoking and survival time of lung cancer patients in a TCGA (The Cancer Genome Atlas) cohort study. The real data application identifies four mediate CpGs, three of which are newly found.

Duration of Intrastate Wars of Attrition: The Causal Impact of Military Build-Ups

Scientific knowledge on the effectiveness of governmental military build-ups to terminate intrastate conflicts is sparse and inconclusive. Developing a war-of-attrition framework, we derive the impact of governments' armaments on the duration of these conflicts: military build-ups, as reflected in inflows of major conventional weapons, enable the government to inflict costs onto its adversaries, forcing them to withdraw earlier from the conflict. This type of weapons is required in particular to project military power over larger distances and to fight rebels in remote areas. Using SIPRI arms transfer data for the first time in a disaggregated dyadic design, covering 418 government-rebel group dyads in 134 conflicts, we empirically corroborate the formal model's predictions. As endogeneity issues might arise when governments procure arms in anticipation of a protracted conflict, we ensure causal identification with an instrumental variable survival approach based on the Aalen additive hazards model.

Variable selection for a mark-specific additive hazards model using the adaptive LASSO

In HIV vaccine efficacy trials, mark-specific hazards models have important applications and can be used to evaluate the strain-specific vaccine efficacy. Additive hazards models have been widely used in practice, especially when continuous covariates are present. In this article, we conduct variable selection for a mark-specific additive hazards model. The proposed method is based on an estimating equation with the first derivative of the adaptive LASSO penalty function. The asymptotic properties of the resulting estimators are established. The finite sample behavior of the proposed estimators is evaluated through simulation studies, and an application to a dataset from the first HIV vaccine efficacy trial is provided.