Standards & Methodology

Every notebook in KASS adheres to specific quality thresholds. These standards aren’t about academic posturing—they’re about producing analysis that decision-makers can actually rely on.

Identification Strategy

Clear statement of what variation identifies the causal effect, what assumptions are required, and what tests validate those assumptions.

Every notebook answers:

• What is the source of identifying variation?
• What assumptions are required for causal interpretation?
• How do we test those assumptions?
• What would invalidate the analysis?

Robustness & Specification Tests

Not just one regression. Alternative specifications, placebo tests, sensitivity analyses, and honest discussion of what the results do and don’t support.

Standard robustness checks include:

• Alternative bandwidths (for RD designs)
• Different functional forms (linear, polynomial, nonparametric)
• Placebo outcomes (variables that shouldn’t be affected)
• Placebo treatments (times/places without treatment)
• Sensitivity to unobservables (Oster bounds, Rosenbaum bounds)

Reproducibility

Complete data pipelines from source to final result. If you can’t reproduce it, it’s not here.

Requirements:

• All data acquisition code included
• Random seeds set for stochastic methods
• Environment specifications provided
• Results regenerate identically

Honest Limitations

Every method has boundaries. We state them explicitly rather than hoping nobody notices.

Each notebook includes:

• External validity: Where do these results generalize?
• Internal validity threats: What could bias the estimates?
• Data limitations: What’s missing or measured with error?
• Policy applicability: What can and can’t this analysis inform?

Audit & Remediation

All applied econometrics notebooks underwent independent audit in January 2026 and meet institutional standards for transparency, reproducibility, and methodological rigor.