Loess API

# loessBandwidth

Loess bandwidth. Thickness for the confidence of the loess curve. A vaulue between 0 and 1.
Default: 0.8

# loessCell

loessCell float
Loess cell. If interpolation is used this controls the accuracy of the approximation via the maximum number of points in a cell in the kd tree. Cells with more than floor(n*span*cell) points are subdivided.
Default: 0.2

# loessDegree


# loessFamily

loessFamily option
Loess Family. Type of error distribution to be optimized in fitting. The default, "gaussian" is fitting with least squares. Fitting with Tukey's biweight M-Estimator is used if family = "symmetric".
Default: gaussian
Options: gaussian, symmetric

# loessIteration


# loessNormalize


# loessSpan

loessSpan float
Loess span. Factor to adjust the gaussian shape of the loess curve. A vaulue between 0 and 1.
Default: 0.75

# loessStatistics

LoessStatistics. The statistics to be computed exactly, approximately or not at all? Exact computation can be very slow
Default: approximate
Options: approximate, exact

# loessSteps


# loessSurface

loessSurface option
Loess Surface. The fitted surface to be computed exactly ("direct") or via interpolation from a kd tree.
Default: interpolate
Options: direct, interpolate

# showLoessFit