Dsensitivity builds a sensitivity function for the D- or Ds-optimality criterion which relies on defaults to speed up evaluation.
FedorovWynn for instance requires this behaviour/protocol.
a vector of names or indices, the subset of parameters to use.
Defaults to the set of parameters in use.
names
a vector of names or indices, the set of parameters to use.
Defaults to the parameters for which the Fisher information is available.
defaults
a named list of default values.
The value NULL is equivalent to absence.
Value
Dsensitivity returns function(x=NULL, desw=NULL, desx=NULL, mod=NULL), the sensitivity function.
It's attributes contain this function's arguments.
Details
For efficiency reasons the returned function won't complain about missing arguments immediately, leading to strange errors.
Please make sure that all arguments are specified at all times.
This behaviour might change in future releases.