incr.conv {coneproj}

Specify an Increasing and Convex Shape-Restriction in a SHAPEREG Formula

Description

A symbolic routine to define that the mean vector is increasing and convex in a predictor in a formula argument to coneproj.

Usage

incr.conv(x, numknots = 0, knots = 0, space = "E")

Arguments

x

A numeric predictor which has the same length as the response vector.

numknots

The number of knots used to smoothly constrain a predictor. The value should be 0 for a shape-restricted predictor without smoothing. The default value is 0.

knots

The knots used to smoothly constrain a predictor. The value should be 0 for a shape-restricted predictor without smoothing. The default value is 0.

space

A character specifying the method to create knots. It will not be used for a shape-restricted predictor without smoothing. The default value is "E".

Details

"incr.conv" returns the vector "x" and imposes on it two attributes: name and shape.

The shape attribute is 5 ("increasing and convex"), and according to the value of the vector itself and this attribute, the cone edges of the cone generated by the constraint matrix, which constrains the relationship between the mean vector and "x" to be increasing and convex, will be made. The cone edges are a set of basis employed in the hinge algorithm.

Note that "incr.conv" does not make the corresponding cone edges itself. It sets things up to a subroutine called makedelta in coneproj.

See references cited in this section for more details.

Value

The vector x with the shape attribute, i.e., shape: 5 ("increasing and convex").

Author(s)

Mary C. Meyer and Xiyue Liao

References

Meyer, M. C. (2013b) A simple new algorithm for quadratic programming with applications in statistics. Communications in Statistics 42(5), 1126–1139.

See Also

incr.conc, incr

Examples

  data(cubic)

  # extract x
  x <- cubic$x

  # extract y
  y <- cubic$y

  # regress y on x with the shape restriction: "increasing" and "convex"
  ans <- shapereg(y ~ incr.conv(x))

  # make a plot
  par(mar = c(4, 4, 1, 1))
  plot(x, y, cex = .7, xlab = "x", ylab = "y")
  lines(x, fitted(ans), col = 2)
  legend("topleft", bty = "n", "shapereg: increasing and convex fit", col = 2, lty = 1)