MachineShop: LARSModel – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

LARSModel

Least Angle Regression, Lasso and Infinitesimal Forward Stagewise Models

Description

Fit variants of Lasso, and provide the entire sequence of coefficients and fits, starting from zero to the least squares fit.

Usage

LARSModel(
  type = c("lasso", "lar", "forward.stagewise", "stepwise"),
  trace = FALSE,
  normalize = TRUE,
  intercept = TRUE,
  step = NULL,
  use.Gram = TRUE
)

Arguments

`type`	model type.
`trace`	logical indicating whether status information is printed during the fitting process.
`normalize`	whether to standardize each variable to have unit L2 norm.
`intercept`	whether to include an intercept in the model.
`step`	algorithm step number to use for prediction. May be a decimal number indicating a fractional distance between steps. If specified, the maximum number of algorithm steps will be `ceiling(step)`; otherwise, `step` will be set equal to the source package default maximum [default: `max.steps`].
`use.Gram`	whether to precompute the Gram matrix.

Details

Response Types:: numeric
Automatic Tuning of Grid Parameters:: step

Default values for the NULL arguments and further model details can be found in the source link below.

Value

MLModel class object.

Examples

## Requires prior installation of suggested package lars to run

fit(sale_amount ~ ., data = ICHomes, model = LARSModel)

MachineShop

Machine Learning Models and Tools

v2.8.0

GPL-3

Authors

Brian J Smith [aut, cre]

Initial release

2021-04-16