ctsem (version 3.0.4)

ctCI: ctCI Computes confidence intervals on specified parameters / matrices for already fitted ctsem fit object.

Description

ctCI Computes confidence intervals on specified parameters / matrices for already fitted ctsem fit object.

Usage

ctCI(ctfitobj, confidenceintervals, optimizer = "NPSOL", verbose = 0)

Arguments

ctfitobj

Already fit ctsem fit object (class: ctsemFit) to estimate confidence intervals for.

confidenceintervals

character vector of matrices and or parameters for which to estimate 95% confidence intervals for.

optimizer

character vector. Defaults to NPSOL (recommended), but other optimizers available within OpenMx (e.g. 'CSOLNP') may be specified.

verbose

Integer between 0 and 3 reflecting amount of output while calculating.

Value

ctfitobj, with confidence intervals included.

Details

Confidence intervals typically estimate more reliably using the proprietary NPSOL optimizer available within OpenMx only when installing directly from OpenMx website. Use command " source('http://openmx.psyc.virginia.edu/getOpenMx.R') " to install OpenMx with NPSOL. If estimating for a multigroup model, specify confidence intervals as normal, e.g. confidenceintervals = c('DRIFT', 'diffusion_Y1_Y1') . The necessary group prefixes are added internally.

Examples

# NOT RUN {
## Examples set to 'donttest' because they take longer than 5s.
# }
# NOT RUN {
data("ctExample3")
model <- ctModel(n.latent = 1, n.manifest = 3, Tpoints = 100, 
 LAMBDA = matrix(c(1, "lambda2", "lambda3"), nrow = 3, ncol = 1), 
 MANIFESTMEANS = matrix(c(0, "manifestmean2", "manifestmean3"), nrow = 3, 
   ncol = 1))
fit <- ctFit(dat = ctExample3, ctmodelobj = model, objective = "Kalman",
 stationary = c("T0VAR"))

fit <- ctCI(fit, confidenceintervals = 'DRIFT')

summary(fit)$omxsummary$CI
# }