ENH: Allow to plot weighted KDEs. (#59337)

* added weights argument in _plot function and modified scipy.stats.gaussian_kde accordingly

* Update pandas/plotting/_core.py

Co-authored-by: Xiao Yuan <[email protected]>

* moved "weights" after "ind"

* added entry in whatsnew v3.0.0.rst

* added new test for weights

* removed weights

* moved message from Plotting section to Other enhancements

---------

Co-authored-by: Xiao Yuan <[email protected]>

Author: fbourgey

doc/source/whatsnew/v3.0.0.rst

@@ -35,6 +35,7 @@ Other enhancements
 - :meth:`DataFrame.agg` called with ``axis=1`` and a ``func`` which relabels the result index now raises a ``NotImplementedError`` (:issue:`58807`).
 - :meth:`Index.get_loc` now accepts also subclasses of ``tuple`` as keys (:issue:`57922`)
 - :meth:`Styler.set_tooltips` provides alternative method to storing tooltips by using title attribute of td elements. (:issue:`56981`)
+- Added missing parameter ``weights`` in :meth:`DataFrame.plot.kde` for the estimation of the PDF (:issue:`59337`)
 - Allow dictionaries to be passed to :meth:`pandas.Series.str.replace` via ``pat`` parameter (:issue:`51748`)
 - Support passing a :class:`Series` input to :func:`json_normalize` that retains the :class:`Series` :class:`Index` (:issue:`51452`)
 - Support reading value labels from Stata 108-format (Stata 6) and earlier files (:issue:`58154`)

pandas/plotting/_core.py

@@ -1450,6 +1450,7 @@ def kde(
         self,
         bw_method: Literal["scott", "silverman"] | float | Callable | None = None,
         ind: np.ndarray | int | None = None,
+        weights: np.ndarray | None = None,
         **kwargs,
     ) -> PlotAccessor:
         """
@@ -1475,6 +1476,9 @@ def kde(
             1000 equally spaced points are used. If `ind` is a NumPy array, the
             KDE is evaluated at the points passed. If `ind` is an integer,
             `ind` number of equally spaced points are used.
+        weights : NumPy array, optional
+            Weights of datapoints. This must be the same shape as datapoints.
+            If None (default), the samples are assumed to be equally weighted.
         **kwargs
             Additional keyword arguments are documented in
             :meth:`DataFrame.plot`.
@@ -1560,7 +1564,7 @@ def kde(
 
             >>> ax = df.plot.kde(ind=[1, 2, 3, 4, 5, 6])
         """
-        return self(kind="kde", bw_method=bw_method, ind=ind, **kwargs)
+        return self(kind="kde", bw_method=bw_method, ind=ind, weights=weights, **kwargs)
 
     density = kde
 

pandas/plotting/_matplotlib/hist.py

@@ -269,6 +269,7 @@ def _plot(  #  type: ignore[override]
         y: np.ndarray,
         style=None,
         bw_method=None,
+        weights=None,
         ind=None,
         column_num=None,
         stacking_id: int | None = None,
@@ -277,7 +278,7 @@ def _plot(  #  type: ignore[override]
         from scipy.stats import gaussian_kde
 
         y = remove_na_arraylike(y)
-        gkde = gaussian_kde(y, bw_method=bw_method)
+        gkde = gaussian_kde(y, bw_method=bw_method, weights=weights)
 
         y = gkde.evaluate(ind)
         lines = MPLPlot._plot(ax, ind, y, style=style, **kwds)

pandas/tests/plotting/test_series.py

@@ -538,6 +538,22 @@ def test_kde_kwargs(self, ts, bw_method, ind):
         pytest.importorskip("scipy")
         _check_plot_works(ts.plot.kde, bw_method=bw_method, ind=ind)
 
+    @pytest.mark.parametrize(
+        "bw_method, ind, weights",
+        [
+            ["scott", 20, None],
+            [None, 20, None],
+            [None, np.int_(20), None],
+            [0.5, np.linspace(-100, 100, 20), None],
+            ["scott", 40, np.linspace(0.0, 2.0, 50)],
+        ],
+    )
+    def test_kde_kwargs_weights(self, bw_method, ind, weights):
+        # GH59337
+        pytest.importorskip("scipy")
+        s = Series(np.random.default_rng(2).uniform(size=50))
+        _check_plot_works(s.plot.kde, bw_method=bw_method, ind=ind, weights=weights)
+
     def test_density_kwargs(self, ts):
         pytest.importorskip("scipy")
         sample_points = np.linspace(-100, 100, 20)