| Number |
CREPEDP-79 |
| Publication Date |
August 2020 |
| Title |
Variable Selection in Double/debiased Machine Learning
for Causal Inference: An Outcome-Adaptive Approach |
| Author |
Daijiro Kabata and Mototsugu Shintani
|
| Abstract |
Access to high-dimensional data has made the use of machine learning in the causal inference more common in recent years. The double/debiased machine learning (DML) estimator for the treatment effect is designed to obtain the valid inference when nuisance functions, in the treatment and the outcome equations, are estimated using machine learning methods. However, when some covariates in the treatment equation are not correlated with the outcome, inclusion of such covariates, called instruments, in the estimation of the propensity score in the treatment equation will result in increasing bias and variance of DML estimator. To solve this issue, we introduce an outcome-adaptive DML estimator which incorporates the outcome-adaptive lasso to exclude the instruments from the propensity score. We evaluate the performance of the proposed method using Monte Carlo simulation. The results indicate that our proposed method outperforms other methods in many cases.
|
| Keywords |
Causal inference, Double/debiased Machine Learning, High-dimensional data, Machine Learning, Outcome-Adaptive Lasso. |
| Other information |
Paper in English (24 pages) |