Lectures - Causal Inference Flashcards
What’s a potential outcome?
What’s causal inference?
What are some population causal estimands for binary treatment?
What’s the fundamental problem of causal inference?
FUNDAMENTAL PROBLEM OF CAUSAL INFERENCE: only observe one potential outcome on each subject and without ridiculously strong assumptions the individual causal efect is not identiied
Causal inference is inherently a missing data problem
What’s interference and SUTVA?
What’s a causal effect and what inferences might be made about such an effect?
What’s a statistical model?
A statistical model is a mathematical model that embodies a set of statistical assumptions concerning the generation of sample data (and similar data from a larger population). A statistical model represents, often in considerably idealized form, the data-generating process.[1] When referring specifically to probabilities, the corresponding term is probabilistic model.
What’s a causal model?
What is meant by “potential outcomes are assumed a baseline covariate”?
How might we show the natural estimator for the average treatment effect (ATE) is consistent?
What’s a natural estimator for the average treatment effect (ATE)?
What’s the problem with observational studies?
Why is missing not at random a problem?
Sample is not representative of population of interest. Standard estimation and inference techniques do not hold.
What’s the strong ignorability/ no unmeasured confounders assumption?
What’s the Regression Estimator?
