docs(sphinx): expose v0.2.0 submodules in API reference + index quickstarts

docs/api.rst:
  Three new automodule sections (Case-Crossover Estimators,
  Model Comparison, Dose-Response and Breakpoint Analysis) with
  one-paragraph leads explaining what each submodule covers.
  Existing sections unchanged.

docs/index.rst:
  Features list rewritten to flag core-vs-v0.2.0 capabilities.
  Three new quickstart code blocks (benjamini_hochberg on a
  heterogeneous battery; broken_stick_fit + davies_test + hill_fit
  on synthetic dose-response data; URL-corrected docs badge).

docs/conf.py:
  release bumped from '0.1.0' to '0.2.0'.

Author: bgeurten

docs/api.rst

@@ -41,6 +41,44 @@ Multi-Group Testing
    :undoc-members:
    :show-inheritance:
 
+Case-Crossover Estimators (v0.2.0)
+----------------------------------
+
+Time-stratified case-crossover conditional logit with stratified-
+permutation backup and Burke-2015 σ-rescaled effect translator.
+
+.. automodule:: rerandomstats.case_crossover
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Model Comparison (v0.2.0)
+-------------------------
+
+Wald two-sample-β test, nested-model likelihood-ratio test, and the
+shared-algorithmic-source p-value correction helpers
+(``correct_pvalues`` / ``correct_pvalues_array`` /
+``benjamini_hochberg``) routing through ``statsmodels.stats.multitest.multipletests``.
+
+.. automodule:: rerandomstats.model_comparison
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
+Dose-Response and Breakpoint Analysis (v0.2.0)
+----------------------------------------------
+
+Broken-stick segmented regression with profile-RSS 95 % CI on the
+breakpoint, the Davies (1987 / 2002) and Muggeo (2016) Pseudo-Score
+breakpoint-existence tests, the 4-parameter Hill / logistic fit with
+Sebaugh–McCray lower-bend point, and a per-subject iterator that
+applies any of the four to a panel of subjects.
+
+.. automodule:: rerandomstats.dose_response
+   :members:
+   :undoc-members:
+   :show-inheritance:
+
 Resampling Index Generator
 --------------------------
 

docs/conf.py

@@ -12,7 +12,7 @@
 project = "reRandomStats"
 copyright = "2023-2026, Bart R.H. Geurten"
 author = "Bart R.H. Geurten"
-release = "0.1.0"
+release = "0.2.0"
 
 # -- General configuration ---------------------------------------------------
 extensions = [

docs/index.rst

@@ -1,30 +1,69 @@
 reRandomStats Documentation
 ============================
 
-.. image:: https://github.com/zerotonin/rerandomstats/actions/workflows/tests.yml/badge.svg
-   :target: https://github.com/zerotonin/rerandomstats/actions/workflows/tests.yml
+.. image:: https://github.com/zerotonin/reRandomStats/actions/workflows/tests.yml/badge.svg
+   :target: https://github.com/zerotonin/reRandomStats/actions/workflows/tests.yml
    :alt: Tests
 
+.. image:: https://github.com/zerotonin/reRandomStats/actions/workflows/docs.yml/badge.svg
+   :target: https://zerotonin.github.io/reRandomStats/
+   :alt: Docs
+
 .. image:: https://img.shields.io/badge/License-MIT-yellow.svg
    :target: https://opensource.org/licenses/MIT
    :alt: License: MIT
 
-A comprehensive Python toolkit for **re-randomisation statistics** in the
-tradition of Sir Ronald A. Fisher.
+A comprehensive Python toolkit for **re-randomisation statistics** in
+the tradition of Sir Ronald A. Fisher, extended in v0.2.0 with
+time-stratified case-crossover estimators, model-comparison helpers
+that share a single algorithmic source for multiple-comparisons
+correction, and a dose-response / breakpoint-analysis toolkit
+(broken-stick + Davies + Muggeo Pseudo-Score + 4-parameter Hill).
 
 Features
 --------
 
-- Fisher's Resampling Test with flexible test statistics (mean, median, sum)
-- Fisher's Exact Test for 2×2 contingency tables
-- Multi-group pairwise comparisons with FDR correction
-- Binomial proportion tests with Wilson confidence intervals
-- Classical hypothesis test dispatcher (Mann-Whitney U, Kruskal-Wallis, etc.)
-- CSV data I/O with German-locale support and wide→long conversion
+Core (v0.1.0+):
+
+- **Fisher's Resampling Test** — flexible test statistics
+  (mean / median / sum differences)
+- **Fisher's Exact Test** — 2×2 contingency table
+- **Multi-Group Pairwise Testing** — with automatic multiple-testing
+  correction
+- **Binomial Proportion Tests** — single- and two-sample variants
+  with Wilson confidence intervals
+- **Classical Hypothesis Tests** — unified dispatcher for
+  Mann-Whitney U, Kruskal-Wallis, Kolmogorov-Smirnov, Mood's
+  Median, Wilcoxon Rank-Sum, independent t-test, chi-square
+- **Data I/O** — CSV reader with German-locale support and
+  wide → long table conversion
+- **Pretty-Table Output** — manuscript-ready ASCII / Markdown tables
+
+New in v0.2.0 — three submodules:
+
+- **case_crossover** — time-stratified case-crossover conditional
+  logit (Maclure 1991, Lee et al. 2023) with stratified-permutation
+  backup, daylight-hours covariate, within-event temporal-contrast
+  test (hot-day-vs-hot-week), and Burke-2015 σ-rescaled effect
+  translator
+- **model_comparison** — two-sample Wald z-test on
+  independently-estimated coefficients, nested-model likelihood-
+  ratio test, single-method correction (`correct_pvalues`), array
+  helper (`correct_pvalues_array`), and dual-method BH + Bonferroni
+  report (`benjamini_hochberg`); all four correction helpers route
+  through one `statsmodels.stats.multitest.multipletests` call
+- **dose_response** — broken-stick segmented regression with
+  profile-RSS 95 % CI on the breakpoint, Davies (1987 / 2002) and
+  Muggeo (2016) Pseudo-Score tests for breakpoint existence,
+  4-parameter Hill fit with Sebaugh–McCray (2003) lower-bend point,
+  and a per-subject iterator that applies any of the four to a panel
+  of subjects (pickle-safe for ``concurrent.futures.ProcessPoolExecutor``)
 
 Quick Start
 -----------
 
+Core resampling test:
+
 .. code-block:: python
 
    from rerandomstats import FisherResamplingTest
@@ -38,6 +77,43 @@ Quick Start
    p_value = test.main()
    print(f"p = {p_value:.4f}")
 
+Multi-comparisons correction on a heterogeneous battery:
+
+.. code-block:: python
+
+   from rerandomstats import benjamini_hochberg
+
+   # p-values from arbitrary tests (case-crossover, Wald, Poisson, LRT, …)
+   result = benjamini_hochberg({
+       "H1": 0.001, "H2": 0.012, "H3": 0.040, "H4": 0.080,
+   })
+   for name, row in result["results"].items():
+       print(f"{name}: q = {row['bh_adjusted_p']:.4f}  "
+             f"BH-reject = {row['bh_reject']}")
+
+Breakpoint detection on dose-response data:
+
+.. code-block:: python
+
+   import numpy as np
+   from rerandomstats import broken_stick_fit, davies_test, hill_fit
+
+   x = np.random.uniform(10, 30, 300)
+   y = np.where(x <= 22, 38 + 0.05 * x,
+                38 + 0.05 * 22 + 0.8 * (x - 22))
+   y += np.random.normal(0, 0.2, 300)
+
+   bs = broken_stick_fit(x, y)
+   print(f"Breakpoint = {bs['breakpoint']:.2f} "
+         f"[{bs['breakpoint_ci_lo']:.2f}, {bs['breakpoint_ci_hi']:.2f}]")
+
+   davies = davies_test(x, y)
+   print(f"Davies p = {davies['pvalue']:.4f}")
+
+   hill = hill_fit(x, y)
+   print(f"EC50 = {hill['ec50']:.2f}, Hill n = {hill['hill_n']:.2f}, "
+         f"lower bend = {hill['lower_bend']:.2f}")
+
 .. toctree::
    :maxdepth: 2
    :caption: Contents