Mon: A causal-inference version of a statistics problem: If you fit a regression model with interactions, and the underlying process has an interaction, your coefficients won’t be directly interpretable.
Tues: He’s looking for probability puzzles
Wed: In criticism of criticism of criticism
Thurs: A question about physics-types models for flows in economics
Fri: What I got wrong (and right) about econometrics and unbiasedness
Sat: Social networks spread disease—but they also spread practices that reduce disease
Sun: Collaborative filtering, hierarchical modeling, and . . . speed dating