Fix passing of numeric_only argument

Author: Arno Veenstra

pandas/core/series.py

@@ -3676,7 +3676,8 @@ def _reduce(self, op, name, axis=0, skipna=True, numeric_only=None,
 
         # dispatch to ExtensionArray interface
         if isinstance(delegate, ExtensionArray):
-            return delegate._reduce(name, skipna=skipna, **kwds)
+            return delegate._reduce(name, skipna=skipna,
+                                    numeric_only=numeric_only, **kwds)
         elif is_datetime64_dtype(delegate):
             # use DatetimeIndex implementation to handle skipna correctly
             delegate = DatetimeIndex(delegate)

pandas/tests/series/test_analytics.py

@@ -1497,3 +1497,11 @@ def test_drop_duplicates_categorical_bool(self, is_ordered):
         sc = tc.copy()
         sc.drop_duplicates(keep=False, inplace=True)
         tm.assert_series_equal(sc, tc[~expected])
+
+    def test_min_numeric_only(self):
+        raw_cat = pd.Categorical(["a", "b", "c", "a"],
+                                 categories=["b", "c", "d"],
+                                 ordered=True)
+        s = pd.Series(raw_cat)
+        result = s.min(numeric_only=True)
+        assert result == 'b'