Repository Mirror for your Cloud Server and Webhosting

Title:

Constructs Principal Surfaces

Version:

1.0

Description:

Construct a principal surface that are two-dimensional surfaces that pass through the middle of a p-dimensional data set. They minimise the distance from the data points, and provide a nonlinear summary of data. The surfaces are nonparametric and their shape is suggested by the data. The formation of a surface is found using an iterative procedure which starts with a linear summary, typically with a principal component plane. Each successive iteration is a local average of the p-dimensional points, where an average is based on a projection of a point onto the nonlinear surface of the previous iteration. For more information on principal surfaces, see Ganey, R. (2019, "https://open.uct.ac.za/items/4e655d7d-d10c-481b-9ccc-801903aebfc8").

License:

MIT + file LICENSE

Encoding:

UTF-8

RoxygenNote:

7.3.1

Imports:

rgl

Suggests:

stats, Matrix, akima, knitr, rmarkdown

VignetteBuilder:

knitr

NeedsCompilation:

Packaged:

2025-03-10 16:26:16 UTC; raeesaganey

Author:

Raeesa Ganey

[aut, cre]

Maintainer:

Raeesa Ganey <Raeesa.ganey@wits.ac.za>

Repository:

CRAN

Date/Publication:

2025-03-12 17:30:12 UTC

Principal Surface

Description

A function to compute principal surfaces based on input data containing continuous variables.

Usage

principal.surface(
  X,
  max.iter = 10,
  alpha = 0.6,
  N = 50,
  print_iterations = FALSE
)

Arguments

X

A data frame or matrix containing continuous variables.

max.iter

Integer. Maximum number of iterations for the principal surface algorithm.

alpha

Numeric. The span argument passed to the loess() function.

N

Integer. The resolution for the interpolated grid surface, creating an N^2 \times p matrix.

print_iterations

Logical. Should the iterations in the principal surface algorithm be printed? Defaults to FALSE.

Value

A list with the following components:

fj.mat: A numeric n \times p matrix of the final principal surface fitted values.
lambda.j: A numeric representation of the samples in two dimensions.

Examples


surface <- principal.surface(iris[,1:3],max.iter = 3)
surface <- principal.surface(iris[1:50,1:3],max.iter = 3)