loss {scR} | R Documentation |
simvcd()
Utility function to define the least-squares loss function to be optimized for simvcd()
loss(h, ngrid, xi, a = 0.16, a1 = 1.2, a11 = 0.14927)
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 |
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. |
A real number giving the estimated value of the MSE given the current guess.
simvcd()
, the user-facing function for simulating VC dimension and risk_bounds()
to generate estimates for xi.