crossval: Uses cross-validation to find a good weighting of...

In arsbar24/reg-fit: Does Some Simple Regularized Fits To Data

Description

Uses cross-validation to find a good weighting of regularization

Usage

crossval(y, x, partitions = 5, degree = 1, lambda = c(0.1, 1, 10))

Arguments

y	Dependent variable
x	Independent variable(s) in array form
partitions	Number of partitions to divide the data into (n for n-fold validation)
degree	Degree of polynomial to fit
lambda	Vector of lambda weightings to try

Details

Splits the data into the number of partitions specified then attempts a fit on each partition, and calculates the error of the fit on the remaining data.

Value

A list of the the best lambda value, and its fit (includes coefficients and error).

Examples

crossval((1:10)^3, 1:10, partitions = 20, degree = 3, lambda = c(0.001, 0.01, 0.1, 1, 10))

arsbar24/reg-fit documentation built on May 14, 2019, 2 a.m.