scR (version 0.4.0)

loss: Utility function to define the least-squares loss function to be optimized for simvcd()

Description

Utility function to define the least-squares loss function to be optimized for simvcd()

Usage

loss(h, ngrid, xi, a = 0.16, a1 = 1.2, a11 = 0.14927)

Value

A real number giving the estimated value of the MSE given the current guess.

Arguments

h

A positive real number giving the current guess at VC dimension

ngrid

Vector of sample sizes for which the bounding function is estimated.

xi

Vector of estimated values of the bounding function, usually obtained from risk_bounds()

a

Scaling coefficient for the bounding function. Defaults to the value given by Vapnik, Levin and Le Cun 1994.

a1

Scaling coefficient for the bounding function. Defaults to the value given by Vapnik, Levin and Le Cun 1994.

a11

Scaling coefficient for the bounding function. Defaults to the value given by Vapnik, Levin and Le Cun 1994.

See Also

simvcd(), the user-facing function for simulating VC dimension and risk_bounds() to generate estimates for xi.