DOC add quantile regression example

mkdocs.yml

@@ -35,6 +35,7 @@ nav:
     - About: index.md
   - Examples:
     - Regression on Workers' Compensation: examples/regression_on_workers_compensation.ipynb
+    - Quantile Regression: examples/quantile_regression.ipynb
   - API Reference: reference/  # defer to gen-files + literate-nav
   - Development: development.md
   - Release Notes: https://github.com/lorentzenchr/model-diagnostics/releases

src/model_diagnostics/calibration/identification.py

@@ -25,8 +25,8 @@ def identification_function(
 ) -> np.ndarray:
     r"""Canonical identification function.
 
-    Identification functions act as generalised residuals. See Notes for further
-    details.
+    Identification functions act as generalised residuals. See [Notes](#notes) for
+    further details.
 
     Parameters
     ----------

src/model_diagnostics/calibration/plots.py

@@ -42,7 +42,7 @@ def plot_reliability_diagram(
     predictions `y_pred` (x-axis).
     The conditional expectation is estimated via isotonic regression (PAV algorithm)
     of `y_obs` on `y_pred`.
-    See Notes for further details.
+    See [Notes](#notes) for further details.
 
     Parameters
     ----------

src/model_diagnostics/scoring/plots.py

@@ -32,7 +32,7 @@ def plot_murphy_diagram(
     over a range of their free parameter `eta`. This shows, if a model dominates all
     others over a wide class of scoring functions or if the ranking is very much
     dependent on the choice of scoring function.
-    See Notes for further details.
+    See [Notes](#notes) for further details.
 
     Parameters
     ----------
@@ -66,7 +66,7 @@ def plot_murphy_diagram(
 
     Notes
     -----
-    For details, refer to [Ehm2015].
+    For details, refer to `[Ehm2015]`.
 
     References
     ----------

src/model_diagnostics/scoring/scoring.py

@@ -1,4 +1,5 @@
-"""The scoring module provides scoring functions and the score decomposition.
+"""The scoring module provides scoring functions, also known as loss functions,
+and a score decomposition.
 Each scoring function is implemented as a class that needs to be instantiated
 before calling the `__call__` methode, e.g. `SquaredError()(y_obs=[1], y_pred=[2])`.
 """
@@ -576,7 +577,7 @@ class ElementaryScore(_BaseScoringFunction):
 
     The elementary scoring function is consistent for the specified `functional` for
     all values of `eta` and is the main ingredient for Murphy diagrams.
-    See Notes for further details.
+    See [Notes](#notes) for further details.
 
     Parameters
     ----------
@@ -599,7 +600,7 @@ class ElementaryScore(_BaseScoringFunction):
     The elementary scoring or loss function is given by
 
     \[
-    S_\eta^h(y, z) = (\mathbf{1}\{\eta \le z\} - \mathbf{1}\{\eta \le y\})
+    S_\eta(y, z) = (\mathbf{1}\{\eta \le z\} - \mathbf{1}\{\eta \le y\})
     V(y, \eta)
     \]