RobustiPy: An efficient next-generation multiversal library with model selection, averaging, resampling, and explainable AI.

Abstract

Scientific inference is undermined by the vast but rarely explored "multiverse" of defensible modeling choices, which can generate results as variable as the phenomena under study. We introduce RobustiPy, an open-source Python library that systematizes multiverse analysis and model uncertainty quantification at scale. RobustiPy unifies bootstrap-based inference, combinatorial specification search, model selection and averaging, joint inference routines, and explainable AI methods within a modular, reproducible framework. Beyond exhaustive specification curves, it supports rigorous out-of-sample validation and provides feature-level explanations for the full-specification predictive model. We demonstrate utility across five simulations and ten high-profile replications spanning economics, sociology, psychology, and medicine, including a re-analysis of widely cited findings with documented discrepancies. Benchmarking on ∼672 million simulated regressions shows that RobustiPy delivers state-of-the-art computational efficiency while expanding transparency. By standardizing and accelerating methods for robustness, RobustiPy transforms how researchers interrogate sensitivity across the analytical multiverse, offering a foundation for reproducible and interpretable computational science.