vector of real estimates to be converted to link values
model
MARK model object used only to obtain model structure/links
etc. If function is being called for model averaged estimates, then any
model in the model list used to construct the estimates is sufficient
parm.indices
index numbers from PIMS for rows in design
matrix(non-simplified indices); x[parm.indices] are computed
vcv.real
v-c matrix for the real parameters
use.mlogits
logical; if FALSE then parameters with mlogit links are
transformed with logit rather than mlogit for creating confidence intervals
for each value
Value
A list with the estimates (link values) and the links that were
used. If vcv.real = TRUE, then the v-c matrix of the links is also
returned.
Details
It has 2 uses both related to model averaged estimates. Firstly, it is used
to transform model averaged estimates so the normal confidence interval can
be constructed on the link values and then back-transformed to real space.
The second function is to enable parametric bootstrapping in which the error
distbution is assumed to be multivariate normal for the link values. From a
single model, the link values are easily constructed from the betas and
design matrix so this function is not needed. But for model averaging there
is no equivalent because the real parameters are averaged over a variety of
models with the same real parameter structure but differing design
structures. This function allows for link values and their var-cov matrix
to be created from the model averaged real estimates.