Skip to contents

Bootstrap Confidence Intervals for Functional Coefficient

Source: R/fregre.R
fregre.bootstrap.ci.Rd

Computes pointwise and simultaneous bootstrap confidence intervals for the functional coefficient beta(t) in a functional linear or logistic model.

Usage

fregre.bootstrap.ci(model, n.boot = 200, alpha = 0.05, seed = 42)

Arguments

model

A fitted object of class 'fregre.lm' or 'fregre.logistic'.

n.boot

Number of bootstrap replicates (default 200).

alpha

Significance level (default 0.05 for 95 percent CI).

seed

Random seed for reproducibility.

Value

An object of class 'fregre.bootstrap.ci' with components:

  • lower, upper: Pointwise CI bounds.

  • center: Original beta(t) estimate.

  • sim.lower, sim.upper: Simultaneous CI bands.

  • n.boot.success: Number of successful bootstrap replicates.

  • argvals: Grid points.

  • alpha: Significance level used.

  • model.class: Class of the original model.

Examples

if (FALSE) { # \dontrun{
fit <- fregre.lm(fdataobj, y, ncomp = 3)
ci <- fregre.bootstrap.ci(fit, n.boot = 200)
plot(ci)
} # }