umx (version 1.9.1)

umxModify: umxModify: Add, set, or drop model paths by label.

Description

umxModify allows you to modify, re-run and summarize an mxModel, all in one line of script. You can add paths, or other model elements, set paths or drop them. As an example, this one-liner drops a path labelled "Cs", and returns the updated model:

Usage

umxModify(lastFit, update = NULL, master = NULL, regex = FALSE,
  free = FALSE, value = 0, freeToStart = NA, name = NULL,
  verbose = FALSE, intervals = FALSE, comparison = FALSE,
  autoRun = TRUE, dropList = "deprecated")

Arguments

lastFit

The mxModel you wish to update and run.

update

What to update before re-running. Can be a list of labels, a regular expression (set regex = TRUE) or an object such as mxCI etc.

master

If you set master, then the labels in update will be equated (slaved) to those provided in master.

regex

Whether or not update is a regular expression (defaults to FALSE). If you provide a string, it over-rides the contents of update, and sets regex to TRUE.

free

The state to set "free" to for the parameters whose labels you specify (defaults to free = FALSE, i.e., fixed)

value

The value to set the parameters whose labels you specify too (defaults to 0)

freeToStart

Whether to update parameters based on their current free-state. free = c(TRUE, FALSE, NA), (defaults to NA - i.e, not checked)

name

The name for the new model

verbose

How much feedback to give

intervals

Whether to run confidence intervals (see mxRun)

comparison

Whether to run umxCompare() after umxRun

autoRun

Whether to run the modified model before returning it (default), or just to modify and return without running.

dropList

(deprecated: use 'update' instead.

Value

- mxModel

Details

fit2 = umxModify(fit1, update = "Cs", name = "newModelName", comparison = TRUE)

Regular expressions are a powerful feature: they let you drop collections of paths by matching patterns fit2 = umxModify(fit1, regex = "C[sr]", name = "drop_Cs_and_Cr", comparison = TRUE)

If you are just starting out, you might find it easier to be more explicit. Like this:

fit2 = omxSetParameters(fit1, labels = "Cs", values = 0, free = FALSE, name = "newModelName") fit2 = mxRun(fit2) summary(fit2)

References

- http://github.com/tbates/umx

Examples

# NOT RUN {
require(umx)
data(demoOneFactor)
latents  = c("G")
manifests = names(demoOneFactor)
umx_set_optimizer("SLSQP")

m1 <- 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(cov(demoOneFactor), type = "cov", numObs = 500)
)
m1 = umxRun(m1, setLabels = TRUE, setValues = TRUE)
m2 = umxModify(m1, update = "G_to_x1", name = "drop_X1")
umxSummary(m2); umxCompare(m1, m2)
# 1-line version including comparison
m2 = umxModify(m1, update = "G_to_x1", name = "drop_X1", comparison = TRUE)
# use regular expression to drop multiple paths: e.g. G to x3, x4, x5
m2 = umxModify(m1, update = "^G_to_x[3-5]", regex = TRUE, name = "no_G_to_x3_5", comp = TRUE)
# Same, but shorter
m2 = umxModify(m1, regex  = "^G_to_x[3-5]", name = "no_G_to_x3_5")
# Same, but don't autoRun
m2 = umxModify(m1, regex  = "^G_to_x[3-5]", name = "no_G_to_x3_5", autoRun = FALSE) 
m2 = umxModify(m1, update = "G_to_x1", value = .2, name = "fix_G_x1_at_point2", comp = TRUE)
m3 = umxModify(m2, update = "G_to_x1", free = TRUE, name = "free_G_x1_again", comparison = TRUE)
# }