mfPole: R6 Class for mfboost pole model object for use in...

In Almond-S/manifoldboost: Boosting Regression Models for Manifold Valued Data

R6 Class for mfboost pole model object for use in mfboost_family

Description

R6 Class for mfboost pole model object for use in mfboost_family

R6 Class for mfboost pole model object for use in mfboost_family

Details

A pole (similar to an offset) for a mfboost model is typically result of some kind of model fit, which might be another simpler mfboost model or some other kind of model procedure estimating the pole from the data. For later model predictions on new data, it is important to store the not only the resulting pole, but the whole model behind it, to have access to new pole predictions. The abstract R6 class mfPole implements the basic structure needed to provide a pole for mfboost.

Public fields

model

A fitted model object.

Methods

Public methods

• mfPole$fit()

• mfPole$predict()

• mfPole$clone()

---

Method fit()

An initialization function determining the pole from the data. The function takes two arguments.

Usage

mfPole$fit(obj.formula, data)

Arguments

obj.formula

Model formula for the pole model.

data

Data in FDboost format.

---

Method predict()

A function taking newdata in the same format as data and generating a prediction as a simple vector fitting to the data.

Usage

mfPole$predict(newdata = NULL)

Arguments

newdata

in FDboost format. The default NULL specifies the original data.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

mfPole$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Almond-S/manifoldboost documentation built on June 23, 2022, 11:06 a.m.