Abstract:
|
In the last decade the machine learning literature has made great advances in predictive modeling in settings with many observations and with many features or covariates. In economics, however, there are many cases where we are not simply interested in predictions, rather we are interested in causal effects, predictions given interventions in the system. In this paper we look at some of the problems in causal modeling where modifications of the machine learning methods may be useful for estimating causal effects.
|