plot.GeDSboost: Plot Method for GeDSboost Objects

In GeDS: Geometrically Designed Spline Regression

Plot Method for GeDSboost Objects

Description

Plots the component functions of a GeDSboost object fitted using NGeDSboost. If the model has a single base-learner, the plot will be returned on the response scale. Otherwise, plots are produced on the linear predictor scale. Note that only univariate base-learner plots are returned, as the representation of the boosted model as a single spline model is available only for univariate base-learners (see Dimitrova et al. (2025)). Additionally, since component-wise gradient boosting inherently performs base-learner selection, plots will only be generated for the base-learners selected during the boosting iterations.

Usage

## S3 method for class 'GeDSboost'
plot(x, n = 3L, ...)

Arguments

x	A GeDSboost object, as returned by NGeDSboost().
n	Integer value (2, 3 or 4) specifying the order (= degree + 1) of the FGB-GeDS fit to be extracted.
...	Further arguments to be passed to the plot.default function.

References

Dimitrova, D. S., Kaishev, V. K. and Saenz Guillen, E. L. (2025). GeDS: An R Package for Regression, Generalized Additive Models and Functional Gradient Boosting, based on Geometrically Designed (GeD) Splines. Manuscript submitted for publication.

See Also

NGeDSboost

Examples

data(mtcars)
# Convert specified variables to factors
categorical_vars <- c("cyl", "vs", "am", "gear", "carb")
mtcars[categorical_vars] <- lapply(mtcars[categorical_vars], factor)
N <- nrow(mtcars); ratio <- 0.8
set.seed(123)
trainIndex <- sample(1:N, size = floor(ratio * N))
# Subset the data into training and test sets
train <- mtcars[trainIndex, ]
test <- mtcars[-trainIndex, ]
Gmodboost <- NGeDSboost(mpg ~ cyl + f(drat) + f(wt) + f(hp) + vs + am,
                        data = train, phi = 0.7, shrinkage = 0.9, initial_learner = FALSE)

par(mfrow = c(2,3))
plot(Gmodboost, n = 2)

GeDS documentation built on June 30, 2025, 9:07 a.m.