Add more documentation and tests

Author: raulcd

cpp/src/arrow/compute/api_aggregate.h

@@ -124,6 +124,9 @@ class ARROW_EXPORT SkewOptions : public FunctionOptions {
   /// If true (the default), null values are ignored. Otherwise, if any value is null,
   /// emit null.
   bool skip_nulls;
+  /// If true (the default), the calculated value is biased. If false, the calculated
+  /// value includes a correction factor to reduce bias, making it more accurate for
+  /// small sample sizes.
   bool bias;
   /// If less than this many non-null values are observed, emit null.
   uint32_t min_count;

cpp/src/arrow/compute/kernels/aggregate_test.cc

@@ -3693,6 +3693,11 @@ TEST_F(TestSkewKurtosis, Options) {
     AssertSkewKurtosisInvalid(type, {"[]", "[]", "[]"}, options);
     AssertSkewKurtosisAre(type, "[0, 1, null, 2]", options, 0.0, -1.5);
     AssertSkewKurtosisAre(type, {"[0, 1]", "[]", "[null, 2]"}, options, 0.0, -1.5);
+    options.bias = false;
+    AssertSkewKurtosisAre(type, {"[0, 1]", "[]", "[null, 2]"}, options, 0.0, -NAN);
+    AssertSkewKurtosisAre(type, {"[1, 2, 3]", "[40, null]"}, options, 1.9889477403978211,
+                          3.9631931024230695);
+    options.bias = true;
     options.min_count = 3;
     AssertSkewKurtosisAre(type, "[0, 1, null, 2]", options, 0.0, -1.5);
     AssertSkewKurtosisAre(type, {"[0, 1]", "[]", "[null, 2]"}, options, 0.0, -1.5);

python/pyarrow/_compute.pyx

@@ -1920,6 +1920,9 @@ class SkewOptions(_SkewOptions):
     Parameters
     ----------
     {_skip_nulls_doc()}
+    bias : bool, default True
+        Whether the calculated value is biased.
+        If False, the value computed includes a corrections factor to reduce bias.
     {_min_count_doc(default=0)}
     """
 

python/pyarrow/tests/test_compute.py

@@ -3841,7 +3841,8 @@ def test_pivot_wider():
 
 
 @pytest.mark.pandas
-def test_biased_skew_and_kurtosis():
+def test_unbiased_skew_and_kurtosis():
+    # Validate computing unbiased skew and kurtosis matches pandas
     input = [1.0, 2.0, 3.0, 40.0, None]
     arrow_skew = pc.skew(input, skip_nulls=True, bias=False)
     pandas_skew = pd.Series(np.array(input)).skew(skipna=True)