Minimise a function subject to linear inequality constraints using an adaptive barrier algorithm.
constrOptim(theta, f, grad, ui, ci, mu = 1e-04, control = list(),
method = if(is.null(grad)) "Nelder-Mead" else "BFGS",
outer.iterations = 100, outer.eps = 1e-05, …,
hessian = FALSE)numeric (vector) starting value (of length \(p\)): must be in the feasible region.
function to minimise (see below).
gradient of f (a function as well),
or NULL (see below).
constraint matrix (\(k \times p\)), see below.
constraint vector of length \(k\) (see below).
(Small) tuning parameter.
passed to optim.
iterations of the barrier algorithm.
non-negative number; the relative convergence tolerance of the barrier algorithm.
Other named arguments to be passed to f and grad:
needs to be passed through optim so should not match its
argument names.
As for optim, but with two extra components:
barrier.value giving the value of the barrier function at the
optimum and outer.iterations gives the
number of outer iterations (calls to optim).
The counts component contains the sum of all
optim()$counts.
The feasible region is defined by ui %*% theta - ci >= 0. The
starting value must be in the interior of the feasible region, but the
minimum may be on the boundary.
A logarithmic barrier is added to enforce the constraints and then
optim is called. The barrier function is chosen so that
the objective function should decrease at each outer iteration. Minima
in the interior of the feasible region are typically found quite
quickly, but a substantial number of outer iterations may be needed
for a minimum on the boundary.
The tuning parameter mu multiplies the barrier term. Its precise
value is often relatively unimportant. As mu increases the
augmented objective function becomes closer to the original objective
function but also less smooth near the boundary of the feasible
region.
Any optim method that permits infinite values for the
objective function may be used (currently all but "L-BFGS-B").
The objective function f takes as first argument the vector
of parameters over which minimisation is to take place. It should
return a scalar result. Optional arguments … will be
passed to optim and then (if not used by optim) to
f. As with optim, the default is to minimise, but
maximisation can be performed by setting control$fnscale to a
negative value.
The gradient function grad must be supplied except with
method = "Nelder-Mead". It should take arguments matching
those of f and return a vector containing the gradient.
K. Lange Numerical Analysis for Statisticians. Springer 2001, p185ff
optim, especially method = "L-BFGS-B" which
does box-constrained optimisation.