basefun (version 0.0-32)

Bernstein_basis: Bernstein Basis Functions

Description

Basis functions defining a Bernstein polynomial

Usage

Bernstein_basis(var, order = 2, ui = c("none", "increasing", "decreasing",  "cyclic", "zerointegral", "positive", "negative"))

Arguments

var

a numeric_var object

order

the order of the polynomial, one defines a linear function

ui

a character describing possible constraints

Details

Bernstein_basis returns a function for the evaluation of the basis functions with corresponding model.matrix and predict methods.

References

Rida T. Farouki (2012), The Bernstein Polynomial Basis: A Centennial Retrospective, Computer Aided Geometric Design, 29(6), 379--419. http://dx.doi.org/10.1016/j.cagd.2012.03.001

Examples

  ### set-up basis
  bb <- Bernstein_basis(numeric_var("x", support = c(0, pi)), 
                        order = 3, ui = "increasing")

  ### generate data + coefficients
  x <- as.data.frame(mkgrid(bb, n = 100))
  cf <- c(1, 2, 2.5, 2.6)

  ### evaluate basis (in two equivalent ways)
  bb(x[1:10,,drop = FALSE])
  model.matrix(bb, data = x[1:10, ,drop = FALSE])

  ### check constraints
  cnstr <- attr(bb(x[1:10,,drop = FALSE]), "constraint")
  all(cnstr$ui %*% cf > cnstr$ci)

  ### evaluate and plot Bernstein polynomial defined by
  ### basis and coefficients
  plot(x$x, predict(bb, newdata = x, coef = cf), type = "l")

  ### evaluate and plot first derivative of 
  ### Bernstein polynomial defined by basis and coefficients
  plot(x$x, predict(bb, newdata = x, coef = cf, deriv = c(x = 1)), 
       type = "l")