umx (version 1.4.0)

umxSummary.MxModel: umxSummary.MxModel

Description

Report the fit of a model in a compact form suitable for a journal. Emits a "warning" when model fit is worse than accepted criterion (TLI >= .95 and RMSEA

Usage

"umxSummary"(model, refModels = NULL, showEstimates = c("raw", "std", "none", "both", "list of column names"), digits = 2, report = c("1", "table", "html"), filter = c("ALL", "NS", "SIG"), SE = TRUE, RMSEA_CI = FALSE, matrixAddresses = FALSE, ...)

Arguments

model

The mxModel whose fit will be reported

refModels

Saturated models if needed for fit indices (see example below: If NULL will be competed on demand. If FALSE will not be computed. Only needed for raw data.

showEstimates

What estimates to show. By default, the raw estimates are shown (Options = c("raw", "std", "both", "none", "list of column names").

digits

How many decimal places to report (default = 2)

report

If "html", then show results in browser ("1", "table", "html")

filter

whether to show significant paths (SIG) or NS paths (NS) or all paths (ALL)

SE

Whether to compute SEs... defaults to TRUE. In rare cases, you might need to turn off to avoid errors.

RMSEA_CI

Whether to compute the CI on RMSEA (Defaults to FALSE)

matrixAddresses

Whether to show "matrix address" columns (Default = FALSE)

...

Other parameters to control model summary

Value

- parameterTable returned invisibly, if estimates requested

Details

notes on CIs and Identification Note, the conventional standard errors reported by OpenMx are used to produce the CIs you see in umxSummary These are used to derive confidence intervals based on the formula 95

Sometimes they appear NA. This often indicates a model which is not identified (seehttp://davidakenny.net/cm/identify.htm). This can include empirical under-identification - for instance two factors that are essentially identical in structure.

A signature of this would be paths estimated at or close to zero. Fixing one or two of these to zero may fix the standard error calculation, and alleviate the need to estimate likelihood-based or bootstrap CIs

If factor loadings can flip sign and provide identical fit, this creates another form of under-identification and can break confidence interval estimation, but I think Fixing a factor loading to 1 and estimating factor variances can help here

References

- Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Coventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

- Yu, C.Y. (2002). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes. University of California, Los Angeles, Los Angeles. Retrieved from http://www.statmodel.com/download/Yudissertation.pdf http://www.github.com/tbates/umx

See Also

- umxRun

Other Reporting functions: RMSEA.MxModel, RMSEA.summary.mxmodel, RMSEA, confint.MxModel, extractAIC.MxModel, loadings, logLik.MxModel, plot.MxModel, residuals.MxModel, umxCI_boot, umxCI, umxCompare, umxExpCov, umxExpMeans, umxFitIndices, umxPlotACEcov, umxPlotACE, umxPlotCP, umxPlotGxE, umxPlotIP, umxSummaryACE, umx_drop_ok, umx_standardize_RAM

Examples

require(umx)
data(demoOneFactor)
latents  = c("G")
manifests = names(demoOneFactor)
m1 <- umxRAM("One Factor",
	data = mxData(cov(demoOneFactor), type = "cov", numObs = 500),
	umxPath(latents, to = manifests),
	umxPath(var = manifests),
	umxPath(var = latents, fixedAt = 1)
)
umxSummary(m1, showEstimates = "std")
# output as latex
umx_set_table_format("latex")
umxSummary(m1, showEstimates = "std")
umx_set_table_format("markdown")
# output as raw
umxSummary(m1, show = "raw")
m1 <- mxModel(m1,
  mxData(demoOneFactor[1:100,], type = "raw"),
  umxPath(mean = manifests),
  umxPath(mean = latents, fixedAt = 0)
)
m1 <- mxRun(m1)
umxSummary(m1, showEstimates = "std", filter = "NS")