Obtain the coefficients, model summary or coefficient variance-covariance
matrix for a model fitted by
1 2 3 4 5 6 7 8
An object of class "PlackettLuce" as returned by
An integer or character string specifying the reference item (for
which log worth will be set to zero). If
A logical indicating whether to return parameters on the log scale
with the item specified by
additional arguments, passed to
By default, parameters are returned on the log scale, as most suited for
log = FALSE, the worth parameters are returned,
constrained to sum to one so that they represent the probability that
the corresponding item comes first in a ranking of all items, given that
first place is not tied.
The variance-covariance matrix is returned for the worth and tie parameters
on the log scale, with the reference as specified by
ref. For models
estimated by maximum likelihood, the variance-covariance is the inverse of
the Fisher information of the log-likelihood.
For models with a normal or gamma prior, the variance-covariance is based on
the Fisher information of the log-posterior. When adherence parameters have
been estimated, the log-posterior is not linear in the parameters. In this
case there is a difference between the expected and observed Fisher
information. By default,
vcov will return the variance-covariance
based on the expected information, but
type gives to option to use
the observed information instead. For large samples, the difference between
these options should be small. Note that the estimation of the adherence
parameters is accounted for in the computation of the variance-covariance
matrix, but only the sub-matrix corresponding to the worth and tie
parameters is estimated.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.