Math::LOESS::Model - Math::LOESS model configurations
version 0.0000_02
You normally don't need to construct object of this class yourself. Instead you get the object from an Math::LOESS object.
The parameter controls the degree of smoothing. Default is 0.75.
For span < 1, the neighbourhood used for the fit includes proportion span of the points, and these have tricubic weighting (proportional to (1 - (dist/maxdist)^3)^3). For span > 1, all points are used, with the "maximum distance" assumed to be span^(1/p) times the actual maximum distance for p explanatory variables.
span
(1 - (dist/maxdist)^3)^3)
span^(1/p)
The degree of the polynomials to be used, normally 1 or 2. Default is 2.
Should any terms be fitted globally rather than locally? Default is false. Terms can be specified by name, number or as a logical vector of the same length as the number of predictors.
For fits with more than one predictor and degree = 2, should the quadratic term be dropped for particular predictors? Default is false. Terms are specified in the same way as for parametric.
Should the predictors be normalized to a common scale if there is more than one? The normalization used is to set the 10% trimmed standard deviation to one. Set to false for spatial coordinate predictors and others known to be on a common scale.
If "gaussian" fitting is by least-squares, and if "symmetric" a re-descending M estimator is used with Tukey's biweight function.
"gaussian"
"symmetric"
Math::LOESS
Stephan Loyd <sloyd@cpan.org>
This software is copyright (c) 2019 by Stephan Loyd.
This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.
To install Math::LOESS, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Math::LOESS
CPAN shell
perl -MCPAN -e shell install Math::LOESS
For more information on module installation, please visit the detailed CPAN module installation guide.