Conditional Distributional Treatment Effect with Kernel Conditional Mean Embeddings and U-Statistic Regression

02/16/2021
by   Junhyung Park, et al.
8

We propose to analyse the conditional distributional treatment effect (CoDiTE), which, in contrast to the more common conditional average treatment effect (CATE), is designed to encode a treatment's distributional aspects beyond the mean. We first introduce a formal definition of the CoDiTE associated with a distance function between probability measures. Then we discuss the CoDiTE associated with the maximum mean discrepancy via kernel conditional mean embeddings, which, coupled with a hypothesis test, tells us whether there is any conditional distributional effect of the treatment. Finally, we investigate what kind of conditional distributional effect the treatment has, both in an exploratory manner via the conditional witness function, and in a quantitative manner via U-statistic regression, generalising the CATE to higher-order moments. Experiments on synthetic, semi-synthetic and real datasets demonstrate the merits of our approach.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

12/07/2020

A U-statistic-based test of treatment effect heterogeneity

Many studies include a goal of determining whether there is treatment ef...
11/09/2018

A Fundamental Measure of Treatment Effect Heterogeneity

In this paper we offer an asymptotically efficient, non-parametric way t...
01/15/2022

Treatment Effect Risk: Bounds and Inference

Since the average treatment effect (ATE) measures the change in social w...
03/05/2021

Estimation of Partially Conditional Average Treatment Effect by Hybrid Kernel-covariate Balancing

We study nonparametric estimation for the partially conditional average ...
10/30/2020

Adapting Neural Networks for Uplift Models

Uplift is a particular case of individual treatment effect modeling. Suc...
03/24/2018

Efficient Discovery of Heterogeneous Treatment Effects in Randomized Experiments via Anomalous Pattern Detection

The randomized experiment is an important tool for inferring the causal ...
03/22/2021

Higher-Order Orthogonal Causal Learning for Treatment Effect

Most existing studies on the double/debiased machine learning method con...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.