CMStatistics 2022: Start Registration
View Submission - CMStatistics
B0498
Title: Disentangling confounding and nonsense associations due to dependence Authors:  Elizabeth Ogburn - Johns Hopkins University (United States) [presenting]
Abstract: Nonsense associations can arise when an exposure and an outcome of interest exhibit similar patterns of dependence. Confounding is present when potential outcomes are not independent of treatment. The purpose is to describe how confusion about these two phenomena results in shortcomings in popular methods in three areas: causal inference with multiple treatments and unmeasured confounding; causal and statistical inference with social network data; and causal inference with spatial data. For each of these three areas, we will demonstrate the flaws in existing methods and describe new methods that were inspired by careful consideration of dependence and confounding.