Get the RMSEA with confidence intervals from model
This function calculates the Root Mean Square Error of the Approximation (RMSEA) for a model and computes confidence intervals for that fit statistic.
omxRMSEA(model, lower=.025, upper=.975, null=.05, ...)
model |
An MxModel object for which the RMSEA is desired |
lower |
The lower confidence bound for the confidence interval |
upper |
The upper confidence bound for the confidence interval |
null |
Value of RMSEA used to test for close fit |
... |
Further named arguments passed to summary |
To help users obtain fit statistics related to the RMSEA, this function confidence intervals and a test for close fit. The user determines how close the fit is required to be by setting the null argument to the value desired for comparison.
A named vector with elements lower, est.rmsea, upper, null, and 'Prob(x <= null)'.
Browne, M. W. & Cudeck, R. (1992). Alternative Ways of Assessing Model Fit. Sociological Methods and Research, 21, 230-258.
require(OpenMx)
data(demoOneFactor)
manifests <- names(demoOneFactor)
latents <- c("G")
factorModel <- mxModel("One Factor",
type="RAM",
manifestVars=manifests,
latentVars=latents,
mxPath(from=latents, to=manifests),
mxPath(from=manifests, arrows=2),
mxPath(from=latents, arrows=2, free=FALSE, values=1.0),
mxData(observed=cov(demoOneFactor), type="cov", numObs=500))
factorRun <- mxRun(factorModel)
factorSat <- mxRefModels(factorRun, run=TRUE)
summary(factorRun, refModels=factorSat)
# Gives RMSEA with 95% confidence interval
omxRMSEA(factorRun, .05, .95, refModels=factorSat)
# Gives RMSEA with 90% confidence interval
# and probability of 'close enough' fitPlease choose more modern alternatives, such as Google Chrome or Mozilla Firefox.