Interface to the internal C++ optimization routine "bfgs"
cppBfgs(
x0,
f_,
min.x = c(-Inf),
max.x = c(+Inf),
prec = 1e-05,
verbose = FALSE
)
the start value
the target function
minimum bound on x (default -Inf
)
maximum bound on x (default +Inf
)
precision (default 1e-5
)
be verbose? (not default)
A list with the following elements:
the minimum abscissa found by the algorithm
the inverse Hessian at par
list of the function evaluation pairs: args
and
vals
the convergence code. 0 is “OK”, -1 is “lost precision”, and +1 is “change not large enough”