Nonparametric Tests of Treatment Effect Homogeneity for Policy-Makers

500.5 KB of files including code and documentation for nonparametric tests of treatment effect heterogeneity. The work by Oliver Dukes, last updated in May 2026, proposes a class of tests for guiding personalized treatment policies. Its utility is demonstrated through simulation studies and a re-analysis of an AIDS clinical trial.

Use Cases

• Testing for quantitative treatment effect heterogeneity based on the proposed nonparametric methods.
• Testing for qualitative treatment effect heterogeneity based on the proposed nonparametric methods.
• Evaluating the population impact of adopting a personalized decision rule versus a rule that discards covariates.
• Guiding treatment policies using tests that incorporate structured assumptions on the conditional average treatment effect.

Strengths

• The tests allow for both continuous and discrete covariates, increasing flexibility.
• The methodology does not require sample splitting to obtain a tractable asymptotic null distribution.
• The proposal is validated through simulation studies and a re-analysis of an AIDS clinical trial.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Oliver Dukes via figshare

Collection Method

Likely contains statistical code and documentation for the proposed methodology.

Freshness

Last updated 2026-05-18 14:03:20; freshness should be verified.