ENH: add percentage threshold to dropna

pandas/core/frame.py

@@ -4874,7 +4874,9 @@ def notna(self) -> "DataFrame":
     def notnull(self) -> "DataFrame":
         return ~self.isna()
 
-    def dropna(self, axis=0, how="any", thresh=None, subset=None, inplace=False):
+    def dropna(
+        self, axis=0, how="any", thresh=None, perc=None, subset=None, inplace=False
+    ):
         """
         Remove missing values.
 
@@ -4904,6 +4906,9 @@ def dropna(self, axis=0, how="any", thresh=None, subset=None, inplace=False):
 
         thresh : int, optional
             Require that many non-NA values.
+        perc : float, optional
+            If a column or row exceeds this percentage threshold of
+            NA-values, it will be dropped.
         subset : array-like, optional
             Labels along other axis to consider, e.g. if you are dropping rows
             these would be a list of columns to include.
@@ -4998,6 +5003,11 @@ def dropna(self, axis=0, how="any", thresh=None, subset=None, inplace=False):
 
         if thresh is not None:
             mask = count >= thresh
+        elif perc is not None:
+            if axis == 0:
+                mask = agg_obj.isna().mean(axis=agg_axis) <= perc
+            else:
+                mask = agg_obj.isna().mean() <= perc
         elif how == "any":
             mask = count == len(agg_obj._get_axis(agg_axis))
         elif how == "all":

pandas/tests/extension/base/missing.py

@@ -134,3 +134,27 @@ def test_use_inf_as_na_no_effect(self, data_missing):
         with pd.option_context("mode.use_inf_as_na", True):
             result = ser.isna()
         self.assert_series_equal(result, expected)
+
+
+def test_dropna_perc():
+    # GH 35299
+    df = pd.DataFrame(
+        {
+            "col": ["A", "A", "B", "B"],
+            "A": [80, np.nan, np.nan, np.nan],
+            "B": [80, np.nan, 76, 67],
+        }
+    )
+
+    # axis = 1
+    expected = pd.DataFrame({"col": ["A", "A", "B", "B"], "B": [80, np.nan, 76, 67]})
+    result = df.dropna(perc=0.5, axis=1)
+    tm.assert_frame_equal(result, expected)
+
+    # axis = 0
+    expected = pd.DataFrame(
+        {"col": ["A", "B", "B"], "A": [80.0, np.nan, np.nan], "B": [80.0, 76.0, 67.0]}
+    )
+    expected.index = [0, 2, 3]
+    result = df.dropna(perc=0.5)
+    tm.assert_frame_equal(result, expected)