Sensitivity Analysis Library in Python. Contains Sobol, Morris, FAST, and other methods.
Sobol sequence generator
Python package and open direction-number data for high-dimensional Sobol’ sequences, including the 50,000-dimensional tkrg-a-ap5 set for quasi-Monte Carlo simulation.
Global sensitivity analysis in JAX: Sobol, Morris, PCE, Shapley and ten more methods, vectorised and differentiable
experiment with the quasi-random Sobol sequence
Python derivatives analytics with Monte Carlo and Sobol pricing, Black-Scholes validation, variance reduction, Greeks, implied volatility, and Streamlit.
Análisis de riesgo operativo de entregas mediante análisis causal y técnicas de sensibilidad global con Sobol, FAST y PAWN en Databricks.
Zero-dependency seeded key generation framework for Python — define parameter spaces, generate unique keys, and persist results.
Constrained non-convex optimization under a hard evaluation budget (Stanford AA222): a five-stage hybrid of penalty-method projected gradient, Sobol quasi-random sampling, and constraint-gradient nudging that returns feasible points on 499+ of 500 random starts, including a 10D problem with 8 constraints
Surrogate + causal-discovery pipeline deriving design rules for PCM-integrated envelopes across eight Iranian climates (EnergyPlus campaign, CatBoost surrogates, Morris/Sobol, PC/GES/NOTEARS, SHAP, symbolic regression)
Reproducibility code and results for “From Efficient Frontier to Fragile Frontier: A Global Sensitivity Analysis of Markowitz Portfolios”
To associate your repository with the sobol topic, visit your repo's landing page and select "manage topics."