Skip to content

Fix rolling var bug #62514

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Oct 4, 2025
Merged

Fix rolling var bug #62514

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
cf9f8fa
fix: bug in sliding window
suzyahyah Sep 30, 2025
ac2dd40
feat: consolidate tests for rolling variance correctness against numpy
suzyahyah Sep 30, 2025
fea0d10
lint: reformatted by pre-commit run
suzyahyah Sep 30, 2025
e4a3939
fix: use hardcoded values for tests
suzyahyah Oct 1, 2025
6431ae5
fix: sliding window exclusive end index
suzyahyah Oct 1, 2025
b0aa633
docs: add changeline to v3.0.0.rst
suzyahyah Oct 4, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion pandas/_libs/window/aggregations.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -442,7 +442,7 @@ def roll_var(const float64_t[:] values, ndarray[int64_t] start,

# Over the first window, observations can only be added
# never removed
if i == 0 or not is_monotonic_increasing_bounds or s >= end[i - 1]:
if i == 0 or not is_monotonic_increasing_bounds or s <= end[i]:

prev_value = values[s]
num_consecutive_same_value = 0
Expand Down
143 changes: 123 additions & 20 deletions pandas/tests/window/test_rolling.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,13 +4,11 @@
)

import numpy as np
from numpy.lib.stride_tricks import sliding_window_view
import pytest

from pandas.compat import (
IS64,
is_platform_arm,
is_platform_power,
is_platform_riscv64,
)
from pandas.errors import Pandas4Warning

Expand Down Expand Up @@ -1082,28 +1080,133 @@ def test_rolling_sem(frame_or_series):
tm.assert_series_equal(result, expected)


@pytest.mark.xfail(
is_platform_arm() or is_platform_power() or is_platform_riscv64(),
reason="GH 38921",
)
@pytest.mark.parametrize(
("func", "third_value", "values"),
("func", "values", "window", "ddof", "exp_value"),
[
("var", 1, [5e33, 0, 0.5, 0.5, 2, 0]),
("std", 1, [7.071068e16, 0, 0.7071068, 0.7071068, 1.414214, 0]),
("var", 2, [5e33, 0.5, 0, 0.5, 2, 0]),
("std", 2, [7.071068e16, 0.7071068, 0, 0.7071068, 1.414214, 0]),
(
"var",
[
2.72993945,
1.58444294,
4.14371708,
4.92961687,
2.7138744,
3.48168586,
0.69505519,
1.87511994,
4.20167276,
0.04797675,
],
3,
1,
"numpy_compute",
),
(
"std",
[
2.72993945,
1.58444294,
4.14371708,
4.92961687,
2.7138744,
3.48168586,
0.69505519,
1.87511994,
4.20167276,
0.04797675,
],
3,
1,
"numpy_compute",
),
(
"var",
[
2.72993945,
1.58444294,
4.14371708,
4.92961687,
2.7138744,
3.48168586,
0.69505519,
1.87511994,
4.20167276,
0.04797675,
],
2,
1,
"numpy_compute",
),
(
"std",
[
2.72993945,
1.58444294,
4.14371708,
4.92961687,
2.7138744,
3.48168586,
0.69505519,
1.87511994,
4.20167276,
0.04797675,
],
2,
1,
"numpy_compute",
),
("var", [99999999999999999, 1, 1, 2, 3, 1, 1], 2, 1, 0),
("std", [99999999999999999, 1, 1, 2, 3, 1, 1], 2, 1, 0),
("var", [99999999999999999, 1, 2, 2, 3, 1, 1], 2, 1, 0),
("std", [99999999999999999, 1, 2, 2, 3, 1, 1], 2, 1, 0),
("var", [1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0], 5, 0, "numpy_compute"),
],
)
def test_rolling_var_numerical_issues(func, third_value, values):
# GH: 37051
ds = Series([99999999999999999, 1, third_value, 2, 3, 1, 1])
result = getattr(ds.rolling(2), func)()
expected = Series([np.nan] + values)
tm.assert_series_equal(result, expected)
def test_rolling_var_correctness(func, values, window, ddof, exp_value):
# This tests subsume the previous tests under test_rolling_var_numerical_issues
# GH: 37051, 42064, 54333
ts = Series(values)
result = getattr(ts.rolling(window=window, center=True), func)(ddof=ddof)
if result.last_valid_index():
result = result[
result.first_valid_index() : result.last_valid_index() + 1
].reset_index(drop=True)
expected = Series(
getattr(sliding_window_view(values, window_shape=window), func)(
axis=-1, ddof=ddof
)
)
tm.assert_series_equal(result, expected, atol=1e-55)
# GH 42064
# new `roll_var` will output 0.0 correctly
tm.assert_series_equal(result == 0, expected == 0)
if exp_value == 0:
# new `roll_var` will output 0.0 correctly
tm.assert_series_equal(result == 0, expected == 0)


def test_rolling_var_numerical_stability():
# GH 52407
A = [
0.00000000e00,
0.00000000e00,
3.16188252e-18,
2.95781651e-16,
2.23153542e-51,
0.00000000e00,
0.00000000e00,
5.39943432e-48,
1.38206260e-73,
0.00000000e00,
]
ts = Series(A)

result = ts.rolling(window=3, center=True).var(ddof=1)
result = result[
result.first_valid_index() : result.last_valid_index() + 1
].reset_index(drop=True)

# numpy implementation
expected = Series(sliding_window_view(A, window_shape=3).var(axis=-1, ddof=1))
tm.assert_series_equal(result, expected, atol=1e-55)


def test_timeoffset_as_window_parameter_for_corr(unit):
Expand Down
Loading