Representation of models of formal class "glimML" fitted by maximum-likelihood method.
Objects can be created by calls of the form new("glimML", ...)
or,
more commonly, via the functions betabin
or negbin
.
CALL
The call of the function.
link
The link function used to transform the mean: “logit”, “cloglog” or “log”.
method
The type of fitted model: “BB” for beta-binomial and “NB” for negative-binomial models.
formula
The formula used to model the mean.
random
The formula used to model the overdispersion parameter \(\phi\).
data
Data set to which model was fitted. Different from the original data in case of missing value(s).
param
The vector of the ML estimated parameters \(b\) and \(\phi\).
varparam
The variance-covariance matrix of the ML estimated parameters \(b\) and \(\phi\).
fixed.param
The vector of the ML estimated fixed-effect parameters \(b\).
random.param
The vector of the ML estimated random-effect (correlation) parameters \(\phi\).
logL
The log-likelihood of the fitted model.
logL.max
The log-likelihood of the maximal model (data).
dev
The deviance of the model, i.e., - 2 * (logL - logL.max)
.
df.residual
The residual degrees of freedom of the fitted model.
nbpar
The number of estimated parameters, i.e., nbpar = total number of parameters - number
of fixed parameters. See argument fixpar
in betabin
or negbin
.
iterations
The number of iterations performed in optim
.
code
An integer (returned by optim
) indicating why the optimization process terminated.
Relative gradient is close to 0, current iterate is probably solution.
Successive iterates within tolerance, current iterate is probably solution.
Last global step failed to locate a point lower than estimate. Either estimate is an approximate
local minimum of the function or steptol
is too small.
Iteration limit exceeded.
Maximum step size stepmax
exceeded 5 consecutive times. Either the function is unbounded below,
becomes asymptotic to a finite value from above in some direction or stepmax
is too small.
msg
Message returned by optim
.
singular.hessian
Logical: true when fitting provided a singular hessian, indicating an overparamaterized model.
param.ini
The initial values provided to the ML algorithm.
na.action
A function defining the action taken when missing values are encountered.