Increasing Power for Observational Studies of Aberrant Response: An Adaptive Approach

07/15/2019
by   Siyu Heng, et al.
0

In many observational studies, the interest is in the effect of treatment on bad, aberrant outcomes rather than the average outcome. For such settings, the traditional approach is to define a dichotomous outcome indicating aberration from a continuous score and use the Mantel-Haenszel test with matched data. For example, studies of determinants of poor child growth use the World Health Organization's definition of child stunting being height-for-age z-score ≤ -2. The traditional approach may lose power because it discards potentially useful information about the severity of aberration. We develop an adaptive approach that makes use of this information and improves power. We show our approach asymptotically dominates the traditional approach and performs well both asymptotically and in simulation studies. We develop our approach in two parts. First, we develop an aberrant rank approach in matched observational studies and prove a novel design sensitivity formula enabling its asymptotic comparison with the Mantel-Haenszel test under various settings. Second, we develop a new, general adaptive approach, the two-stage programming method, and use it to adaptively combine the aberrant rank test and the Mantel-Haenszel test. We apply our approach to a study of the effect of teenage pregnancy on stunting.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/18/2019

The uniform general signed rank test and its design sensitivity

A sensitivity analysis in an observational study tests whether the quali...
research
09/12/2023

Sensitivity Analysis for Quantiles of Hidden Biases in Matched Observational Studies

In matched observational studies, the inferred causal conclusions preten...
research
07/11/2022

Covariate-adaptive randomization inference in matched designs

It is common to conduct causal inference in matched observational studie...
research
12/19/2018

Multivariate one-sided testing in matched observational studies as an adversarial game

We present a multivariate one-sided sensitivity analysis for matched obs...
research
04/23/2018

Is My Matched Dataset As-If Randomized, More, Or Less? Unifying the Design and Analysis of Observational Studies

Matching alleviates the problem of covariate imbalance in observational ...
research
01/08/2020

Stratified Pilot Matching in R: The stratamatch Package

In a block-randomized controlled trial, individuals are subdivided by pr...
research
11/29/2018

Adaptive Sparse Estimation with Side Information

The article considers the problem of estimating a high-dimensional spars...

Please sign up or login with your details

Forgot password? Click here to reset