In WuhaoStatistic/bonuslab: lab4

Description

This package is for bonus lab. The main content is about ridge regression.In this vignette,there will be comparison between results from this package and results from "lm.ridge" in "MASS" package.

Import

    library(bonuslab)
    library(MASS)

Methods

The example below is to introduce how these functions work ridgereg() takes three parameter, object : a formula object ; data : a data.frame object lambda : a number

QR() function is to calculate coefficients.

We can also set a lm.ridge object to start compare.

rid <- ridgereg(Petal.Length~Sepal.Width+Sepal.Length, data=iris,1)
rid$QR()
massrid <- lm.ridge(Petal.Length~Sepal.Width+Sepal.Length, data=iris,lambda = 1)

The print() prints out the vector containing coefficients.

rid$print()

The pred() returns the predicted values.

head(rid$pred())

The coef() returns the coefficients as a named vector.In this section, we will do a comparison.

# this answer comes from my onw ridge regression object
rid$coef()
# following coefficient comes from MASS package lm.ridge
massrid

As we can see,there is only little difference between them.According to "bonuslab.pdf", this difference is because of applying different calculating methods.

WuhaoStatistic/bonuslab documentation built on Dec. 18, 2021, 7:20 p.m.