AntonAntonov/ MonadicQuantileRegression

Software monad for Quantile Regression workflows

Contributed by: Anton Antonov

A software monad implementation for Quantile Regression workflows. B-spline, Chebyshev, and custom function bases are supported.

Installation Instructions

To install this paclet in your Wolfram Language environment, evaluate this code:
PacletInstall["AntonAntonov/MonadicQuantileRegression"]


To load the code after installation, evaluate this code:
Needs["AntonAntonov`MonadicQuantileRegression`"]

Details

The curves and (hyper-)surfaces found by Quantile Regression (QR) are called "regression quantiles".
For two-dimensional data QR is done with a B-spline basis, a Chebyshev functions basis, or a custom function basis.
For higher dimensional data QR is doen with a Non-Uniform Rational B-Spline (NURBS) basis or a custom function basis.
QR is used to find "contextual outliers".
Using the found regression quantiles a conditional distribution estimate can be found at each regressor point.
The found regression quantiles can be used to generate random data that closely resembles the original data.
The software monad has functions for computing linear regession, neural nets regression, and QR.

Paclet Guide