Changelog

All notable changes to this project are documented here.

0.1.0 (Unreleased)

Added

• Linear quantile regression (QuantileRegressor)
• Barrodale-Roberts simplex solver ("br") with rank-inversion CI
• Frisch-Newton interior-point solver ("fn" / "fnb")
• Preprocessing + Frisch-Newton solver ("pfn") for large datasets
• L1-penalised Lasso solver ("lasso")

• Extensible solver registry with get_solver() / register_solver()

• Inference

• summary() with 5 SE methods: rank, iid, nid, ker, boot

• bootstrap() with 3 strategies: xy-pair, wild, mcmb

• Nonparametric estimation (QuantizationQuantileEstimator)

• CLVQ optimal quantization grid construction
• Voronoi cell assignment and conditional quantile estimation

• Bootstrap averaging over multiple grids

• sklearn compatibility

• Full check_estimator compliance (46/46 for linear, 52/52 for nonparametric)
• Pipeline, cross-validation, and grid search support

• BaseQuantileEstimator abstract base class

• Datasets

• load_engel() — Engel food expenditure (235 obs, 1 predictor)

• load_barro() — Barro economic growth (161 obs, 13 predictors)

• Documentation

• GitHub Pages site with MkDocs Material
• Theory guides, API reference, and example notebooks