reg.lm: Fits a polynomial of specified degree to the data with a...

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

Description

Fits a polynomial of specified degree to the data with a regularization term

Usage

reg.lm(y, x, degree = 1, lambda = 1)

Arguments

y	Dependent variable
x	Independent variable(s) in array form
degree	Degree of polynomial to fit
lambda	Weight of regularization term (may be selected using crossval)

Details

Finds the polynomial of specified degree that minimizes

({y}-p({x}))^2+lambda*sum(coeff^2)

Where p is the polynomial and coeff is its coefficients.

This function is useful when concerns about overfitting are present. lambda may be chosen by the crossval function.

Value

List with the coefficients for each term in the polynomial and the error.

Examples

reg.lm((1:10)^3, 1:10, degree = 3, lambda = 1)

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