equality_restriction-class: Class "equality\_restriction"

In FAiR: Factor Analysis in R

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

Author(s)

Ben Goodrich

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)

FAiR documentation built on May 29, 2017, 6:08 p.m.