Activity Number:
|
158
- SPEED: Statistical Methods, Computing, and Applications Part 2
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 8, 2022 : 10:30 AM to 11:15 AM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #323750
|
|
Title:
|
Double Machine Learning in a Semiparametric Approach: An Innovative Causal Inference for Observational Studies
|
Author(s):
|
Lynda Aouar*
|
Companies:
|
University of Northern Colorado
|
Keywords:
|
double machine learning;
semiparametric;
regularization bias;
over-fitting bias;
Neyman orthogonality;
sample-splitting
|
Abstract:
|
Lynda Aouar
Abstract
Double machine learning in a semi-parametric approach has been considered as a powerful tool to infer causality within the observational research framework. In the existence of high dimensional nuisance data and low dimensional targeted parameters, the double machine learning can control for the confounders and estimate the targeted variable of interest that causes the response variable. Using classical methods we will encounter two serious forms of biasedness, the regularization bias, and the over-fitting bias. Double machine learning can get a rid of these two biases using the Neyman orthogonality and moment conditions, and sample splitting respectively. An empirical example of real data such as pollution is used to apply double machine learning results and determine inference about the the causal effect on this observational study.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2022 program
|