Series.describe returns first and last for tz-aware datetimes (#21332)

* Series.describe returns first and last for tz-aware datetimes

GH issue 21328

* parameterize tests

* parameterize names

* use tz_naive_fixture and fix top

* add tz describe test for df

* add bugfix to whatsnew

* fix formatting in whatsnew and add issue number to tests

* final bugfix

Author: louispotok

doc/source/whatsnew/v0.24.0.txt

@@ -317,6 +317,7 @@ Timezones
 - Bug in :class:`Series` constructor which would coerce tz-aware and tz-naive :class:`Timestamp`s to tz-aware (:issue:`13051`)
 - Bug in :class:`Index` with ``datetime64[ns, tz]`` dtype that did not localize integer data correctly (:issue:`20964`)
 - Bug in :class:`DatetimeIndex` where constructing with an integer and tz would not localize correctly (:issue:`12619`)
+- Fixed bug where :meth:`DataFrame.describe` and :meth:`Series.describe` on tz-aware datetimes did not show `first` and `last` result (:issue:`21328`)
 
 Offsets
 ^^^^^^^

pandas/core/generic.py

@@ -20,7 +20,7 @@
     is_bool_dtype,
     is_categorical_dtype,
     is_numeric_dtype,
-    is_datetime64_dtype,
+    is_datetime64_any_dtype,
     is_timedelta64_dtype,
     is_datetime64tz_dtype,
     is_list_like,
@@ -8531,12 +8531,13 @@ def describe_categorical_1d(data):
             if result[1] > 0:
                 top, freq = objcounts.index[0], objcounts.iloc[0]
 
-                if is_datetime64_dtype(data):
+                if is_datetime64_any_dtype(data):
+                    tz = data.dt.tz
                     asint = data.dropna().values.view('i8')
                     names += ['top', 'freq', 'first', 'last']
-                    result += [tslib.Timestamp(top), freq,
-                               tslib.Timestamp(asint.min()),
-                               tslib.Timestamp(asint.max())]
+                    result += [tslib.Timestamp(top, tz=tz), freq,
+                               tslib.Timestamp(asint.min(), tz=tz),
+                               tslib.Timestamp(asint.max(), tz=tz)]
                 else:
                     names += ['top', 'freq']
                     result += [top, freq]

pandas/tests/frame/test_analytics.py

@@ -417,6 +417,28 @@ def test_describe_timedelta_values(self):
                     "max           5 days 00:00:00         0 days 05:00:00")
         assert repr(res) == exp_repr
 
+    def test_describe_tz_values(self, tz_naive_fixture):
+        # GH 21332
+        tz = tz_naive_fixture
+        s1 = Series(range(5))
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s2 = Series(date_range(start, end, tz=tz))
+        df = pd.DataFrame({'s1': s1, 's2': s2})
+
+        expected = DataFrame({'s1': [5, np.nan, np.nan, np.nan, np.nan, np.nan,
+                                     2, 1.581139, 0, 1, 2, 3, 4],
+                              's2': [5, 5, s2.value_counts().index[0], 1,
+                                     start.tz_localize(tz),
+                                     end.tz_localize(tz), np.nan, np.nan,
+                                     np.nan, np.nan, np.nan, np.nan, np.nan]},
+                             index=['count', 'unique', 'top', 'freq', 'first',
+                                    'last', 'mean', 'std', 'min', '25%', '50%',
+                                    '75%', 'max']
+                             )
+        res = df.describe(include='all')
+        tm.assert_frame_equal(res, expected)
+
     def test_reduce_mixed_frame(self):
         # GH 6806
         df = DataFrame({

pandas/tests/series/test_analytics.py

@@ -336,6 +336,23 @@ def test_describe(self):
                           index=['count', 'unique', 'top', 'freq'])
         tm.assert_series_equal(result, expected)
 
+    def test_describe_with_tz(self, tz_naive_fixture):
+        # GH 21332
+        tz = tz_naive_fixture
+        name = tz_naive_fixture
+        start = Timestamp(2018, 1, 1)
+        end = Timestamp(2018, 1, 5)
+        s = Series(date_range(start, end, tz=tz), name=name)
+        result = s.describe()
+        expected = Series(
+            [5, 5, s.value_counts().index[0], 1, start.tz_localize(tz),
+             end.tz_localize(tz)
+             ],
+            name=name,
+            index=['count', 'unique', 'top', 'freq', 'first', 'last']
+        )
+        tm.assert_series_equal(result, expected)
+
     def test_argsort(self):
         self._check_accum_op('argsort', check_dtype=False)
         argsorted = self.ts.argsort()