engineAEFA: estimate full-information item factor analysis models with...

In seonghobae/kaefa: kwangwoon automated exploratory factor analysis

estimate full-information item factor analysis models with combinating random effects

Description

estimate full-information item factor analysis models with combinating random effects

Usage

engineAEFA(data, model = 1, GenRandomPars = T, NCYCLES = 4000,
  BURNIN = 1500, SEMCYCLES = 1000, covdata = NULL, fixed = c(~1,
  ~0, ~-1), random = list(~1 | items), key = NULL,
  accelerate = "squarem", symmetric = F, resampling = T,
  samples = 5000, printDebugMsg = F, fitEMatUIRT = F,
  ranefautocomb = T, tryLCA = T, forcingMixedModelOnly = F,
  forcingQMC = F, turnOffMixedEst = F, anchor = NULL,
  skipggumInternal = F, powertest = F, idling = 60, leniency = F)

Arguments

data	insert data.frame object.
model	specify the mirt model if want to calibrate. accepting mirt::mirt.model object.
GenRandomPars	Try to generate Random Parameters? Default is TRUE
NCYCLES	N Cycles of Robbin Monroe stage (stage 3). Default is 4000.
BURNIN	N Cycles of Metro-hastings burnin stage (stage 1). Default is 1500.
SEMCYCLES	N Cycles of Metro-hastings burnin stage (stage 2). Default is 1000.
covdata	insert covariate data frame where use to fixed and random effect term. if not inserted, ignoring fixed and random effect estimation.
fixed	a right sided R formula for specifying the fixed effect (aka 'explanatory') predictors from covdata and itemdesign.
random	a right sided formula or list of formulas containing crossed random effects of the form v1 + ... v_n | G, where G is the grouping variable and v_n are random numeric predictors within each group. G may contain interaction terms, such as group:items to include cross or person-level interactions effects.
key	item key vector of multiple choices test.
accelerate	a character vector indicating the type of acceleration to use. Default is 'squarem' for the SQUAREM procedure (specifically, the gSqS3 approach)
symmetric	force S-EM/Oakes information matrix to be symmetric? Default is FALSE to detect solutions that have not reached the ML estimate.
resampling	Do you want to do resampling with replace? default is TRUE and activate nrow is over samples argument.
samples	specify the number samples with resampling. default is 5000.
printDebugMsg	Do you want to see the debugging messeages? default is FALSE
fitEMatUIRT	Do you want to fit the model with EM at UIRT? default is FALSE
ranefautocomb	Do you want to find global-optimal random effect combination? default is TRUE
tryLCA	Do you want to try calibrate LCA model if avaliable? default is TRUE
forcingMixedModelOnly	Do you want to forcing the Mixed model calibration? default is FALSE
forcingQMC	Do you want to forcing the use QMC estimation instead MHRM? default is FALSE
turnOffMixedEst	Do you want to turn off mixed effect (multilevel) estimation? default is FALSE
anchor	Set the anchor item names If you want to consider DIF detection. default is NULL.
skipggumInternal	Set the skipping ggum fitting procedure to speed up. default is FALSE.
powertest	Set power test mode. default is FALSE.
idling	Set seconds to idle. default is 60.
leniency	skip second order test. default is FALSE

Value

possible optimal combinations of models in list

Examples

## Not run: 
testMod1 <- engineAEFA(mirt::Science, model = 1)


## End(Not run)

seonghobae/kaefa documentation built on May 8, 2022, 3:31 p.m.