getGrad: Calculate the gradient for the non-linear part of a DRMod object
Description
Calculate the gradient for the non-linear part of a DRMod object. The
gradient is used to calculate variance covariance matrix, confidence
intervals and confidence intervals for predictions, based on a
linearization of the non-linear model see Seber and Wild (2003) for
details. This function is also used to calculate the design criterion
for the calcOptDesign function (which depends on the
asymptotic covariance matrix).
Usage
getGrad(object, dose, uGrad = NULL)
Arguments
object
A DRMod object
dose
Dose where to evaluate the gradient
uGrad
Function to return the gradient of a user defined model, see
Examples of the fitDRModel function.
Value
Matrix containing the gradient
References
Seber, G.A.F and Wild, C.F. (2003). Nonlinear Regression, Wiley & Sons