FAiR (version 0.4-15)

equality_restriction-class: Class "equality\_restriction"

Description

It is not necessary to understand this help page if one merely wants to estimate a factor analysis model. This help page is intended for those who want to modify or extend FAiR or otherwise want some idea of how FAiR works ``behind the scenes''.

This class defines an equality restriction for use in a semi-exploratory or confirmatory factor analysis model.

Objects from the Class

Objects can be created by calls of the form new("equality_restriction", ...). However, rarely if ever, would a user need to construct an object this way. The make_restrictions methods do so internally.

Slots

free:

an integer indicating which cell of a primary pattern matrix is considered “free”

fixed:

an integer vector indicating which cells of a primary pattern matrix are constrained to be equal to the “free” cell

dims:

an integer vector of length two indicating the dimensions of the primary pattern matrix under consideration

rownames:

a character vector of rownames for the primary pattern matrix under consideration

level:

either 1L or 2L indicating the level of the factor analysis model under consideration

Methods

show

signature(object = "equality_restriction"): Prints the equality restriction to the screen

See Also

restrictions.1storder-class, make_restrictions

Examples

showClass("equality_restriction")

man  <- make_manifest(covmat = ability.cov)

## Here is how to set up an equality restriction in a SEFA model the hard way
beta <- matrix(NA_real_, nrow = nrow(cormat(man)), ncol = 2)
rownames(beta) <- rownames(cormat(man))
beta[6,2] <- Inf # "fix" beta[6,2] to Inf provisionally

## Note 11L corresponds to beta[5,2] and 12L corresponds to beta[6,2]
ER   <- new("equality_restriction", free = 11L, fixed = 12L, dims = dim(beta),
            rownames = rownames(cormat(man)), level = 1L)
free <- is.na(beta) # i.e. all but beta[6,2]
beta <- new("parameter.coef.SEFA", x = beta, free = free, num_free = sum(free),
            equalities = list(ER))

Phi  <- diag(2)
free <- lower.tri(Phi)
Phi  <- new("parameter.cormat", x = Phi, free = free, num_free = sum(free))

## Now set up a restriction to prohibit the equal coefficients from being zero
blockers <- matrix(FALSE, nrow = nrow(cormat(man)), ncol = 2)
blockers[5:6,2] <- TRUE # these two cells are hence not allowed to both be zero

res <- make_restrictions(manifest = man, beta = beta, Phi = Phi, 
                         discrepancy = "MLE", criteria = list("block_1st"),
                         methodArgs = list(blockers = blockers))

## Not run: 
# ## Here is the easy way to do the same thing, in the following pop-up menus
# ## select the options to impose equality restrictions and inequality 
# ## restrictions (block those coefficients from being zero in a SEFA model)
# res <- make_restrictions(manifest = man, factors = 2, model = "SEFA")
# ## End(Not run)
show(res)