Set control parameters for loess
fits.
loess.control(surface = c("interpolate", "direct"),
statistics = c("approximate", "exact", "none"),
trace.hat = c("exact", "approximate"),
cell = 0.2, iterations = 4, iterTrace = FALSE, …)
should the fitted surface be computed exactly
("direct"
) or via interpolation from a kd tree? Can be abbreviated.
should the statistics be computed exactly, approximately or not at all? Exact computation can be very slow. Can be abbreviated.
Only for the (default) case (surface =
"interpolate", statistics = "approximate")
: should the trace of
the smoother matrix be computed exactly or approximately? It is recommended to use the approximation
for more than about 1000 data points. Can be abbreviated.
if interpolation is used this controls the accuracy of the
approximation via the maximum number of points in a cell in the kd
tree. Cells with more than floor(n*span*cell)
points are subdivided.
the number of iterations used in robust fitting,
i.e. only if family
is "symmetric"
.
logical (or integer) determining if tracing
information during the robust iterations (iterations
\(\ge
2\)) is produced.
further arguments which are ignored.
A list with components