The main driver behind this talk is the increased use of live trading experiments and AB testing in trading applications due to potential causal biases in quantitative models. This presentation has two objectives. First, it introduces the Mathematical Theory of causal inference. Causal inference is shown to be a natural extension of standard Bayesian probability theory. Second, it motivates and solves various applications of causal inference and causal machine learning to trading. The most direct applications are to Transaction Cost Analysis (TCA) and the estimation of price impact in the presence of an alpha signal.