An Empirical Study of Adaptive Kernel Density Estimation in Detecting Distributional Overlap

Bachelor thesis (2023)

Authors

C. Chen Electrical Engineering, Mathematics and Computer Science

Contributors

J.H. Krijthe Pattern Recognition and Bioinformatics - EEMCS (supervisor 1)
Faculty
Electrical Engineering, Mathematics and Computer Science
Copyright: © 2023 Chao Chen
Kernel Density Estimation Positivity Overlap
More Info
expand_more

Published Date

28-06-2023

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Abstract

Given data from an observational study or a randomized experiment, the positivity assumption must hold in order to draw causal relations between the treatment and outcome. However, there is shortage of automatic tools that verify compliance with the positivity assumption. We present tools that uses adaptive and standard kernel density estimation (KDE) methods for validating the assumption. Our empirical analysis of the methods offers insight into when a KDE method can and can not reliably be applied to verify positivity in datasets.