Version that complies with more general requirements

Version that complies with more general requirements

rq.fit.sfn(
  X,
  y,
  tau = 0.5,
  weights = NULL,
  rhs = (1 - tau) * c(t(X) %*% rep(1, length(y))),
  control,
  lambda,
  ...
)

Arguments

X

structure of the design matrix X stored in csr format

y

response vector

tau

target quantile

weights

optional vector of weights

rhs

right hand size of dual problem

control

control parameters for fitting routines

lambda

ignored

...

other arguments, ignored