밤에 쓰는 편지 밤에 쓰는 편지

https://www.bradyneal.com/causal-inference-course#course-textbook

https://www.bradyneal.com/causal-inference-course#course-textbook

https://www.bradyneal.com/causal-inference-course#course-textbookLast week we saw that without any parametric assumptions, we can't identify the causal graph. If we're willing to make the faithfulness assumption, we can discover up to the Markov equivalence class which in the general case of a class of graphs. It's not going to be a single graph. We're not going to be able to identify the exact ..

https://www.bradyneal.com/causal-inference-course#course-textbookThroughout this book, we have done causal inference, assuming we know the causal graph. What if we don't know the graph? Can we learn it? As you might expect, based on this being a running theme in this book, it will depend on what assumptions we are willing to make. We will refer to this problem as structure identification, which ..

https://www.bradyneal.com/causal-inference-course#course-textbookHow can we identify causal effects when we are in the presence of unobserved confounding? One popular way is to find and use instrumental variables. An instrument (instrumental variable) Z has three key qualities.  It affects on treatment T, it affects Y only through T, and the effect of Z on Y is unconfounded. We depict these qual..

https://www.bradyneal.com/causal-inference-course#course-textbookAll of the methods in Chapter 7 assume that we don't have any unobserved confounding. However, unconfoundedness is an untestable assumption. In observational studies, there could also be some unobserved confounder(s). Therefore, we'd like to know how robust our estimates are to unobserved confounding. The first way we can do is by ..

https://www.bradyneal.com/causal-inference-course#course-textbookOnce we identify some causal estimand by reducing it to a statistical estimand, we still have more work to do. We need to get a corresponding estimate. In this chapter, we'll cover a variety of estimators that we can use to do this. 7.1. PreliminariesWe denote the individual treatment effect (ITE) with τi and average treatment effe..

https://www.bradyneal.com/causal-inference-course#course-textbookIn Section 4.4, we saw that satisfying the backdoor criterion is sufficient to give us identifiability, but is the backdoor criterion also necessary? In other words, is it possible to get identifiability without being able to block all backdoor paths? As an example, consider that we have data generated according to the graph in Fig..