The indirect effect of an exposure on an outcome through an intermediate variable can be identified by a product of regression coefficients under certain causal and regression modeling assumptions. Thus, the null hypothesis of no indirect effect is a composite null hypothesis, as the null holds if either regression coefficient is zero. A consequence is that existing hypothesis tests are either severely underpowered near the origin (i.e., when both coefficients are small with respect to standard errors) or do not preserve type 1 error uniformly over the null hypothesis space. We propose hypothesis tests that (i) preserve level alpha type 1 error, (ii) meaningfully improve power when both true underlying effects are small relative to sample size, and (iii) preserve power when at least one is not. One approach gives a closed-form test that is minimax optimal with respect to local power over the alternative parameter space. Another uses sparse linear programming to produce an approximately optimal test for a Bayes risk criterion. We provide an R package that implements the minimax optimal test.
翻译:根据某些因果和回归模型假设,可以通过回归系数的产物确定通过中间变量接触结果的间接影响。因此,不产生间接影响的无效假设是一种综合的无效假设,如果任何一种回归系数为零,则无效即为无效,其后果是,现有的假设试验要么在来源地附近严重不足(即两个系数相对于标准误差而言都很小),要么没有统一保存在无效假设空间上的第1类错误。我们提议进行假设试验,以便(一) 保持阿尔法第1类的误差,(二) 在真实基本效应与样本大小相比小的情况下,明显提高功率,以及(三) 在至少一个不发生误差时,保留功率。一种方法是对替代参数空间的当地功率进行最优的封闭式试验。另一种是使用稀疏线性编程,以产生对海湾风险标准最优的大致测试。我们提供一套R包,以实施最优试验。