agentsclimarketplace
Sign in

Running placebo analysis

Skill brycewang-stanford/Auto-Empirical-Research-Skills/skills/51-pymc-labs-CausalPy/skills/running-placebo-analysis

Performs placebo-in-time sensitivity analysis with hierarchical null model and optional Bayesian assurance. Use when checking model robustness, verifying lack of pre-intervention effects, or estimating study power.From its SKILL.md

Install
npx -y skills add brycewang-stanford/Auto-Empirical-Research-Skills --skill running-placebo-analysis

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

SKILL.md

1.7 KB, 318 tokens by cl100k_base, as published. Nobody here has run it

Running Placebo Analysis

Executes placebo-in-time sensitivity analysis using the core PlaceboInTime check. Builds a hierarchical Bayesian model of the "status quo" (no-effect) distribution, then compares the actual intervention effect against that learned null. Optionally computes Bayesian assurance (operating characteristics).

Workflow

  1. Fit your experiment: Run a CausalPy experiment (ITS, SC) with a PyMC model.
  2. Configure the check: Create a PlaceboInTime with n_folds, optional experiment_factory, and optional assurance parameters.
  3. Run: Call .run(experiment) (standalone) or use within a Pipeline + SensitivityAnalysis.
  4. Evaluate: Inspect the null distribution (theta_new), p_effect_outside_null, and optional assurance results.

Key Concepts

  • Placebo-in-time: Simulating an intervention at a time when none occurred to check if the model falsely detects an effect.
  • Hierarchical null model: A Bayesian model fitted on fold-level summaries that characterises the distribution of effects under no intervention.
  • Assurance: Bayesian operating characteristics — the probability of correctly detecting a real effect given your expected-effect prior and ROPE.
  • Factory Pattern: Decouples the placebo logic from the specific CausalPy experiment type.

References

What ships with it: 1 file

6.8 KB alongside SKILL.md

reference/

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.