pc: pc

In kader: Kernel Adaptive Density Estimation and Regression

Description

Coefficient p_c of eq. (15.15) in Eichner (2017).

Usage

pc(cc)

Arguments

cc	Numeric vector.

Details

p_c = 1/5 * (3c^2 - 5) / (3 - c^2) * c^2.

For further details see p. 297 f. in Eichner (2017) and/or Eichner & Stute (2013).

Value

Vector of same length and mode as cc.

Note

p_c should be undefined for c = √ 3, but pc is here implemented to return Inf in each element of its return vector for which the corresponding element in cc contains R's value of sqrt(3).

Examples

c0 <- expression(sqrt(5/3))
c1 <- expression(sqrt(3) - 0.01)
cgrid <- seq(1.325, 1.7, by = 0.025)
cvals <- c(eval(c0), cgrid, eval(c1))

plot(cvals, pc(cvals), xaxt = "n", xlab = "c", ylab = expression(p[c]))
axis(1, at = cvals, labels = c(c0, cgrid, c1), las = 2)

kader documentation built on May 1, 2019, 10:13 p.m.