LOESS#

The LOESS transform (LOcally Estimated Scatterplot Smoothing) uses a locally-estimated regression to produce a trend line. LOESS performs a sequence of local weighted regressions over a sliding window of nearest-neighbor points. For standard parametric regression options, see the Regression.

Here is an example of using LOESS to smooth samples from a Gaussian random walk:

import altair as alt
import pandas as pd
import numpy as np

np.random.seed(42)

df = pd.DataFrame({
    'x': range(100),
    'y': np.random.randn(100).cumsum()
})

chart = alt.Chart(df).mark_point().encode(
    x='x',
    y='y'
)

chart + chart.transform_loess('x', 'y').mark_line()

Transform Options#

The transform_loess() method is built on the LoessTransform class, which has the following options:

Click to show table

Property	Type	Description
as	array(`FieldName`)	The output field names for the smoothed points generated by the loess transform. Default value: The field names of the input x and y values.
bandwidth	`number`	A bandwidth parameter in the range `[0, 1]` that determines the amount of smoothing. Default value: `0.3`
groupby	array(`FieldName`)	The data fields to group by. If not specified, a single group containing all data objects will be used.
loess	`FieldName`	The data field of the dependent variable to smooth.
on	`FieldName`	The data field of the independent variable to use a predictor.