Remove how keyword from df.rolling() etc.

doc/source/computation.rst

@@ -253,12 +253,6 @@ accept the following arguments:
   result is NA)
 - ``center``: boolean, whether to set the labels at the center (default is False)
 
-.. warning::
-
-   The ``freq`` and ``how`` arguments were in the API prior to 0.18.0 changes. These are deprecated in the new API. You can simply resample the input prior to creating a window function.
-
-   For example, instead of ``s.rolling(window=5,freq='D').max()`` to get the max value on a rolling 5 Day window, one could use ``s.resample('D').max().rolling(window=5).max()``, which first resamples the data to daily data, then provides a rolling 5 day window.
-
 We can then call methods on these ``rolling`` objects. These return like-indexed objects:
 
 .. ipython:: python

doc/source/whatsnew/v0.22.0.txt

@@ -237,8 +237,8 @@ Removal of prior version deprecations/changes
 - The ``SparseList`` class has been removed (:issue:`14007`)
 - The ``pandas.io.wb`` and ``pandas.io.data`` stub modules have been removed (:issue:`13735`)
 - ``Categorical.from_array`` has been removed (:issue:`13854`)
-- The ``freq`` parameter has been removed from the ``rolling``/``expanding``/``ewm`` methods of DataFrame
-  and Series (deprecated since v0.18). Instead, resample before calling the methods. (:issue:18601)
+- The ``freq`` and ``how`` parameters have been removed from the ``rolling``/``expanding``/``ewm`` methods of DataFrame
+  and Series (deprecated since v0.18). Instead, resample before calling the methods. (:issue:18601 & :issue:18668)
 - ``DatetimeIndex.to_datetime``, ``Timestamp.to_datetime``, ``PeriodIndex.to_datetime``, and ``Index.to_datetime`` have been removed (:issue:`8254`, :issue:`14096`, :issue:`14113`)
 
 .. _whatsnew_0220.performance:

pandas/core/window.py

@@ -106,17 +106,17 @@ def validate(self):
             raise ValueError("closed must be 'right', 'left', 'both' or "
                              "'neither'")
 
-    def _convert_freq(self, how=None):
+    def _convert_freq(self):
         """ resample according to the how, return a new object """
 
         obj = self._selected_obj
         index = None
         return obj, index
 
-    def _create_blocks(self, how):
+    def _create_blocks(self):
         """ split data into blocks & return conformed data """
 
-        obj, index = self._convert_freq(how)
+        obj, index = self._convert_freq()
         if index is not None:
             index = self._on
 
@@ -196,7 +196,7 @@ def _get_index(self, index=None):
             return index, index.asi8
         return index, index
 
-    def _prep_values(self, values=None, kill_inf=True, how=None):
+    def _prep_values(self, values=None, kill_inf=True):
 
         if values is None:
             values = getattr(self._selected_obj, 'values', self._selected_obj)
@@ -320,22 +320,10 @@ def aggregate(self, arg, *args, **kwargs):
     agg = aggregate
 
     _shared_docs['sum'] = dedent("""
-    %(name)s sum
-
-    Parameters
-    ----------
-    how : string, default None
-        .. deprecated:: 0.18.0
-           Method for down- or re-sampling""")
+    %(name)s sum""")
 
     _shared_docs['mean'] = dedent("""
-    %(name)s mean
-
-    Parameters
-    ----------
-    how : string, default None
-        .. deprecated:: 0.18.0
-           Method for down- or re-sampling""")
+    %(name)s mean""")
 
 
 class Window(_Window):
@@ -549,17 +537,14 @@ def _pop_args(win_type, arg_names, kwargs):
             # GH #15662. `False` makes symmetric window, rather than periodic.
             return sig.get_window(win_type, window, False).astype(float)
 
-    def _apply_window(self, mean=True, how=None, **kwargs):
+    def _apply_window(self, mean=True, **kwargs):
         """
         Applies a moving window of type ``window_type`` on the data.
 
         Parameters
         ----------
         mean : boolean, default True
             If True computes weighted mean, else weighted sum
-        how : string, default to None
-            .. deprecated:: 0.18.0
-               how to resample
 
         Returns
         -------
@@ -569,7 +554,7 @@ def _apply_window(self, mean=True, how=None, **kwargs):
         window = self._prep_window(**kwargs)
         center = self.center
 
-        blocks, obj, index = self._create_blocks(how=how)
+        blocks, obj, index = self._create_blocks()
         results = []
         for b in blocks:
             try:
@@ -686,7 +671,7 @@ def __init__(self, obj, *args, **kwargs):
     cov = GroupByMixin._dispatch('cov', other=None, pairwise=None)
 
     def _apply(self, func, name, window=None, center=None,
-               check_minp=None, how=None, **kwargs):
+               check_minp=None, **kwargs):
         """
         dispatch to apply; we are stripping all of the _apply kwargs and
         performing the original function call on the grouped object
@@ -710,7 +695,7 @@ def _constructor(self):
         return Rolling
 
     def _apply(self, func, name=None, window=None, center=None,
-               check_minp=None, how=None, **kwargs):
+               check_minp=None, **kwargs):
         """
         Rolling statistical measure using supplied function. Designed to be
         used with passed-in Cython array-based functions.
@@ -723,9 +708,6 @@ def _apply(self, func, name=None, window=None, center=None,
         window : int/array, default to _get_window()
         center : boolean, default to self.center
         check_minp : function, default to _use_window
-        how : string, default to None
-            .. deprecated:: 0.18.0
-               how to resample
 
         Returns
         -------
@@ -739,7 +721,7 @@ def _apply(self, func, name=None, window=None, center=None,
         if check_minp is None:
             check_minp = _use_window
 
-        blocks, obj, index = self._create_blocks(how=how)
+        blocks, obj, index = self._create_blocks()
         index, indexi = self._get_index(index=index)
         results = []
         for b in blocks:
@@ -803,7 +785,7 @@ class _Rolling_and_Expanding(_Rolling):
 
     def count(self):
 
-        blocks, obj, index = self._create_blocks(how=None)
+        blocks, obj, index = self._create_blocks()
         index, indexi = self._get_index(index=index)
 
         window = self._get_window()
@@ -849,45 +831,30 @@ def sum(self, *args, **kwargs):
 
     _shared_docs['max'] = dedent("""
     %(name)s maximum
+    """)
 
-    Parameters
-    ----------
-    how : string, default 'max'
-        .. deprecated:: 0.18.0
-           Method for down- or re-sampling""")
-
-    def max(self, how=None, *args, **kwargs):
+    def max(self, *args, **kwargs):
         nv.validate_window_func('max', args, kwargs)
-        return self._apply('roll_max', 'max', how=how, **kwargs)
+        return self._apply('roll_max', 'max', **kwargs)
 
     _shared_docs['min'] = dedent("""
     %(name)s minimum
+    """)
 
-    Parameters
-    ----------
-    how : string, default 'min'
-        .. deprecated:: 0.18.0
-           Method for down- or re-sampling""")
-
-    def min(self, how=None, *args, **kwargs):
+    def min(self, *args, **kwargs):
         nv.validate_window_func('min', args, kwargs)
-        return self._apply('roll_min', 'min', how=how, **kwargs)
+        return self._apply('roll_min', 'min', **kwargs)
 
     def mean(self, *args, **kwargs):
         nv.validate_window_func('mean', args, kwargs)
         return self._apply('roll_mean', 'mean', **kwargs)
 
     _shared_docs['median'] = dedent("""
     %(name)s median
+    """)
 
-    Parameters
-    ----------
-    how : string, default 'median'
-        .. deprecated:: 0.18.0
-           Method for down- or re-sampling""")
-
-    def median(self, how=None, **kwargs):
-        return self._apply('roll_median_c', 'median', how=how, **kwargs)
+    def median(self, **kwargs):
+        return self._apply('roll_median_c', 'median', **kwargs)
 
     _shared_docs['std'] = dedent("""
     %(name)s standard deviation
@@ -1709,23 +1676,20 @@ def aggregate(self, arg, *args, **kwargs):
 
     agg = aggregate
 
-    def _apply(self, func, how=None, **kwargs):
+    def _apply(self, func, **kwargs):
         """Rolling statistical measure using supplied function. Designed to be
         used with passed-in Cython array-based functions.
 
         Parameters
         ----------
         func : string/callable to apply
-        how : string, default to None
-            .. deprecated:: 0.18.0
-               how to resample
 
         Returns
         -------
         y : type of input argument
 
         """
-        blocks, obj, index = self._create_blocks(how=how)
+        blocks, obj, index = self._create_blocks()
         results = []
         for b in blocks:
             try:

pandas/tests/test_window.py

@@ -3019,7 +3019,7 @@ def test_rolling_max_gh6297(self):
         x = series.resample('D').max().rolling(window=1).max()
         tm.assert_series_equal(expected, x)
 
-    def test_rolling_max_how_resample(self):
+    def test_rolling_max_resample(self):
 
         indices = [datetime(1975, 1, i) for i in range(1, 6)]
         # So that we can have 3 datapoints on last day (4, 10, and 20)
@@ -3040,17 +3040,17 @@ def test_rolling_max_how_resample(self):
         # Now specify median (10.0)
         expected = Series([0.0, 1.0, 2.0, 3.0, 10.0],
                           index=[datetime(1975, 1, i, 0) for i in range(1, 6)])
-        x = series.resample('D').median().rolling(window=1).max(how='median')
+        x = series.resample('D').median().rolling(window=1).max()
         tm.assert_series_equal(expected, x)
 
         # Now specify mean (4+10+20)/3
         v = (4.0 + 10.0 + 20.0) / 3.0
         expected = Series([0.0, 1.0, 2.0, 3.0, v],
                           index=[datetime(1975, 1, i, 0) for i in range(1, 6)])
-        x = series.resample('D').mean().rolling(window=1).max(how='mean')
+        x = series.resample('D').mean().rolling(window=1).max()
         tm.assert_series_equal(expected, x)
 
-    def test_rolling_min_how_resample(self):
+    def test_rolling_min_resample(self):
 
         indices = [datetime(1975, 1, i) for i in range(1, 6)]
         # So that we can have 3 datapoints on last day (4, 10, and 20)
@@ -3068,7 +3068,7 @@ def test_rolling_min_how_resample(self):
         r = series.resample('D').min().rolling(window=1)
         tm.assert_series_equal(expected, r.min())
 
-    def test_rolling_median_how_resample(self):
+    def test_rolling_median_resample(self):
 
         indices = [datetime(1975, 1, i) for i in range(1, 6)]
         # So that we can have 3 datapoints on last day (4, 10, and 20)