Function to control estimation of smooth offset
o_control(k_min = 20, rule = 2, silent = TRUE, cyclic = FALSE, knots = NULL)
a list with controls
maximal number of k in s()
which rule to use in approx() of the response before calculating the
global mean, rule=1 means no extrapolation, rule=2 means to extrapolate the
closest non-missing value, see approx
print error messages of model fit?
defaults to FALSE, if TRUE cyclic splines are used
arguments knots passed to gam