http://seanjtaylor.github.io/CausalInference/
5th year PhD candidate at NYU Stern
Dependence is useful. It's how we build predictive models.
Notice how a dependency doesn't say which direction the causal relationship is.
Example: McDonald's opening doesn't cause obesity, it causes overeating... which causes obesity
Example: High natural ability causes good SAT scores and success in college.
Example: Your car won't start either because you ran out of gas or the battery is dead. These are independent events, but if you condition on the car not starting, they are anti-correlated.
http://idlewords.com/2010/03/scott_and_scurvy.htm
Estimate: \( \Pr(Y_i \mid X_i) \)
Predict: \( \Pr(Y_i \mid X_i = 1) \)
Estimate: \( \Pr(Y_i \mid do(X_i = 1)) \neq \Pr(Y_i \mid X_i = 1) \)
Set: \( X_i = 1 \) for some units \( i \).
Causality is easy in natural sciences.
Molecules, cells, animals, plants are exchangeable!
There is always a great deal we don't observe about people.
\( Y_i = \beta X_i + \epsilon_i \)
Often assumes:
\( X_i \rightarrow Y_i \leftarrow \epsilon_i \)
\( \beta \) is usually a biased estimate of the causal effect.
This is because \( \epsilon_i \) is not exogenous!
Often easy to think of ways that \( \epsilon_i \rightarrow X_i \).
You can always argue that there exists some component of \( \epsilon_i \) that affects \( X_i \) and driving the outcome.
\( Y_i(0) \) is the outcome of \( i \) under no treatment.
\( Y_i(1) \) is the outcome of \( i \) under treatment.
Problem: we can only ever observe one of these values per row.
Key idea: enforce that \( X_i \) is exogenous (has no parents) by assigning it randomly
Prevents any confounding from being possible.
The gold standard in clinical trials and policy experiments.
Key idea: make comparisons between observations that are as similar as possible on a boatload of observable dimensions.
“Social Influence Bias: A Randomized Experiment”
“Distinguishing influence-based contagion from homophily-driven diffusion in dynamic networks”
\( X_i \) : number of adopter friends
\( Y_i \) : whether the user adopts
The problem:
All treated adopters (filled circles) and the number of treated adopters that can be explained by homophily (open circles) per day and cumulatively over time.g