optimize_lambda: Optimize Lambda for a given ridge regression function

In Zebedial/bis557: The package created for BIS557 Homework 1

Description

This function help to choose optimal lambda parameter from a given set by conducting cross-validation to the models with different lambdas and return the lambda that gives the smallest mean squared error. This function was based on the function ridge_regression.

Usage

optimize_lambda(formula, data, lambda.set, folds = 10, ...)

Arguments

formula	a symbolic description of the model to be fitted. This should be a formula class argument.
data	Specification of a dataframe that contains the variables in the model.
lambda.set	a set of lambdas to be checked. This should be a vector of values or a sequence of numbers.
folds	Number of folds used in the k-fold cross-validation
...	other inputs that need to be passed to the function ridge_regression.

Value

a optimal value from the specified set of lambda which returns the smallest MSE in the cross-validation check, the vector of MSE over all lambdas, and the minimum of the MSE.

Examples

data(iris)
optimize_lambda(Sepal.Length ~ ., iris, seq(0,2,0.01))

Zebedial/bis557 documentation built on Dec. 21, 2020, 2:16 a.m.