given correlation. How can we be sure we have covered them all? …(中略)…We would not know what to test for.(Sociology 8th, Giddens,2011) 出典︓ja.wikipedia.org/wiki/アンソニーギデンズ Anthony Giddens
∈ {0, 1} Potential Outcomes: • Y(0) - random variable giving outcome received in the control state • Y(1) - random variable giving outcome received in the treatment state (Only one of Y(0) or Y(1) can ever be observed - the other is a counterfactual outcome) Causal effect: ![# 1 − # 0 ] ATE(Average Treatment Effect)という