Causal Inference

Causal inference is essential for informed decision-making, as it uncovers the true data-generating processes beyond mere associations found in predictive models. The fundamental challenge in causal effect estimation is that we see the outcome under the treatment a sample received. We do not see the outcome that sample would have had under the other treatment. Short recap from \[1/2\] Estimating a Treatment Effect with Causal Inference: Conditional average treatment effect (CATE) \(\tau(x) = \mathbb{E}[Y(1) - Y(0) \mid x]\)....

Intro Machine learning models predict by the correlation between features and labels. For an outcome under a different treatment, that correlation can give the wrong answer. Consider the following example. We have three samples: one got treatment \(T=0\), and two got treatment \(T=1\). sample treatment \(T\) Observed outcome \(Y\) \(Y(0)\), had they received \(T=0\) \(Y(1)\), had they received \(T=1\) 1 0 9 9 12 2 1 3 0 3 3 1 3 0 3 A naive way to measure whether treatment \(T=1\) or \(T=0\) is more helpful:...