Member-only story
The other day I was looking over the Journal of Machine Learning Research, searching for an interesting topic that I could explore, and I came across an article on double machine learning. I had never heard of double machine learning before, so I decided to research it.
What is double machine learning?
Double machine learning is a statistical method used for estimating causal effects in econometrics and other fields. The approach combines two machine learning models to address the potential confounding effect of covariates, or other variables that may affect the outcome being studied.
In double machine learning, one machine learning model is used to estimate the treatment effect, while the other model is used to estimate the conditional expectation of the outcome given the covariates. The two models are combined in a way that allows for consistent estimation of the causal effect, even when the covariates are high-dimensional and potentially correlated with the treatment variable.
The double machine learning approach has been shown to provide accurate and efficient estimates of causal effects, and is increasingly being used in empirical research across a range of disciplines.
