Implementation:Online ml River Stats EWVar

Knowledge Sources	Online_ml_River
Domains	Online_Learning, Statistics
Last Updated	2026-02-08 16:00 GMT

Overview

EWVar computes the exponentially weighted variance of a data stream.

Description

This statistic calculates the exponentially weighted moving variance using the relationship Var(X) = Mean(x^2) - Mean(x)^2. It maintains exponentially weighted means of both x and x^2 internally to compute the variance. The fading factor parameter controls how quickly old observations are discounted. The implementation is optimized using Rust for performance.

Usage

Use EWVar when you need to track the variance of a stream while giving more importance to recent observations. This is useful for monitoring volatility in financial data, detecting changes in noise levels, quality control, and anomaly detection where recent variance patterns are more relevant than historical ones.

Code Reference

Source Location

• Repository: Online_ml_River
• File: river/stats/ewvar.py

Signature

class EWVar(stats.base.Univariate):
    def __init__(self, fading_factor=0.5):
        if not 0 <= fading_factor <= 1:
            raise ValueError("q is not comprised between 0 and 1")
        self.fading_factor = fading_factor
        self._ewvar = _rust_stats.RsEWVar(fading_factor)

Import

from river import stats

I/O Contract

Inputs

Name	Type	Required	Description
x	numbers.Number	Yes	Value to update the statistic with
fading_factor	float	Yes (init)	Fading factor between 0 and 1 (default: 0.5)

Outputs

Name	Type	Description
get()	float	Current exponentially weighted variance

Usage Examples

from river import stats

# Create exponentially weighted variance with fading factor 0.5
ewv = stats.EWVar(fading_factor=0.5)

X = [1, 3, 5, 4, 6, 8, 7, 9, 11]

for x in X:
    ewv.update(x)
    print(f"Value: {x}, EW Variance: {ewv.get():.4f}")

# Output:
# Value: 1, EW Variance: 0.0000
# Value: 3, EW Variance: 1.0000
# Value: 5, EW Variance: 2.7500
# Value: 4, EW Variance: 1.4375
# Value: 6, EW Variance: 1.9844
# Value: 8, EW Variance: 3.4336
# Value: 7, EW Variance: 1.7959
# Value: 9, EW Variance: 2.1990
# Value: 11, EW Variance: 3.5654

# Use with EWMean for complete statistics
ewm = stats.EWMean(fading_factor=0.5)
ewv = stats.EWVar(fading_factor=0.5)

for x in [10, 15, 12, 18, 20]:
    ewm.update(x)
    ewv.update(x)
    print(f"Mean: {ewm.get():.2f}, Variance: {ewv.get():.2f}")

Related Pages

• Environment:Online_ml_River_Python_Runtime_Environment