ENH: Implement FloatingArray reductions

pandas/conftest.py

@@ -1063,6 +1063,17 @@ def any_nullable_int_dtype(request):
     return request.param
 
 
+@pytest.fixture(params=tm.FLOAT_EA_DTYPES)
+def any_nullable_float_dtype(request):
+    """
+    Parameterized fixture for any nullable float dtype.
+
+    * 'Float32'
+    * 'Float64'
+    """
+    return request.param
+
+
 @pytest.fixture(params=tm.SIGNED_EA_INT_DTYPES)
 def any_signed_nullable_int_dtype(request):
     """

pandas/core/arrays/floating.py

@@ -475,10 +475,19 @@ def _reduce(self, name: str, skipna: bool = True, **kwargs):
 
     def sum(self, skipna=True, min_count=0, **kwargs):
         nv.validate_sum((), kwargs)
-        result = masked_reductions.sum(
-            values=self._data, mask=self._mask, skipna=skipna, min_count=min_count
-        )
-        return result
+        return self._reduce("sum", skipna=skipna, min_count=min_count)
+
+    def prod(self, skipna=True, min_count=0, **kwargs):
+        nv.validate_prod((), kwargs)
+        return self._reduce("prod", skipna=skipna, min_count=min_count)
+
+    def min(self, skipna=True, **kwargs):
+        nv.validate_min((), kwargs)
+        return self._reduce("min", skipna=skipna)
+
+    def max(self, skipna=True, **kwargs):
+        nv.validate_max((), kwargs)
+        return self._reduce("max", skipna=skipna)
 
     def _maybe_mask_result(self, result, mask, other, op_name: str):
         """

pandas/tests/arrays/floating/test_function.py

@@ -152,3 +152,28 @@ def test_preserve_dtypes(op):
         index=pd.Index(["a", "b"], name="A"),
     )
     tm.assert_frame_equal(result, expected)
+
+
+@pytest.mark.parametrize("skipna", [True, False])
+@pytest.mark.parametrize("method", ["min", "max"])
+def test_floating_array_min_max(skipna, method, any_nullable_float_dtype):
+    dtype = any_nullable_float_dtype
+    arr = pd.array([0.0, 1.0, None], dtype=dtype)
+    func = getattr(arr, method)
+    result = func(skipna=skipna)
+    if skipna:
+        assert result == (0 if method == "min" else 1)
+    else:
+        assert result is pd.NA
+
+
+@pytest.mark.parametrize("skipna", [True, False])
+@pytest.mark.parametrize("min_count", [0, 9])
+def test_floating_array_prod(skipna, min_count, any_nullable_float_dtype):
+    dtype = any_nullable_float_dtype
+    arr = pd.array([1.0, 2.0, None], dtype=dtype)
+    result = arr.prod(skipna=skipna, min_count=min_count)
+    if skipna and min_count == 0:
+        assert result == 2
+    else:
+        assert result is pd.NA