update.probit: update method for probit-class
In bomarkussen/probit: Item response analysis with random effects.

Description Usage Arguments Value

View source: R/update_probit.R

update method for probit-object as generated by probit.

## S3 method for class 'probit'
update(
  object,
  fixed = NULL,
  random = NULL,
  A = NULL,
  dependence = NULL,
  data = NULL,
  B = NULL,
  BB = NULL,
  maxit = 20,
  sig.level = 0.6,
  verbose = 0,
  estimate.models = TRUE
)

`object`	Object of class `probit`.
`fixed`	Update of right hand side of model formula for fixed effects. Defaults to `fixed=NULL`, which corresponds to no update.
`random`	List of updates of right hand side of formulas for the random effects. Defaults to `random=NULL`, which corresponds to no update.
`A`	Transformation matrix of predicted random effects. Defaults to `A=NULL`, which corresponds to identity transformation of residing random effects.
`dependence`	Text string (`"marginal" or "joint"`) deciding whether random effects are assumed independent or with a common joint normal distribution. Defaults to `dependence=NULL`, which corresponds to no update.
`data`	Date frame with data on the wide format. Defaults to `data=NULL`, which corresponds to `data` taken from the call object.
`B`	Number of simulations in minimization step. Defaults to `B=NULL`, which corresponds to `B` taken from the call object.
`BB`	Number of simulations per subject in `anova` or `update`. Defaults to `BB=NULL`, which corresponds to `BB` taken from the call object.
`maxit`	Maximal number of minimization-maximization steps. Defaults to `maxit=20`.
`sig.level`	Significance level at which the iterative stochastic optimization will be stopped. Defaults to `sig.level=0.60`.
`verbose`	Numeric controlling amount of convergence diagnostics. Default: `verbose=0` corresponding to no output.
`estimate.models`	Boolean deciding if model parameters are estimated in `update`.