
Minimizes soft insensitive loss function (SILF) for support vector regression.
iSolver(z, a, extra)
Vector containing the fitted values
Vector with Lagrange multipliers
Value of the target function
Gradient at point x
Vector containing observed response
Matrix with active constraints
List with element y
containing the observed response vector, weights
with optional observation weights, beta
between 0 and 1, and eps
> 0
This function is called internally in activeSet
by setting mySolver = iSolver
.
Efron, B. (1991). Regression percentiles using asymmetric squared error loss. Statistica Sinica, 1, 93-125.
activeSet
##Fitting isotone regression using active set
set.seed(12345)
y <- rnorm(9) ##response values
w <- rep(1,9) ##unit weights
eps <- 2
beta <- 0.4
btota <- cbind(1:8, 2:9) ##Matrix defining isotonicity (total order)
fit.silf <- activeSet(btota, iSolver, weights = w, y = y, beta = beta, eps = eps)
Run the code above in your browser using DataLab