efaModel function

Generates an EFA model to be used by lavaan and regsem Function created by Florian Scharf for the paper Should regularization replace simple structure rotation in Exploratory Factor Analysis -- Scharf & Nestler (in press at SEM)

efaModel(nFactors, variables)

Arguments

• nFactors: Number of latent factors to generate.
• variables: Names of variables to be used as indicators

Returns

model Full EFA model parameters.

Examples

## Not run:

HS <- data.frame(scale(HolzingerSwineford1939[,7:15]))
# Note to find number of factors, recommended to use
# fa.parallel() from the psych package
# using the wrong number of factors can distort the results
mod = efaModel(3, colnames(HS))

semFit = sem(mod, data = HS, int.ov.free = FALSE, int.lv.free = FALSE,
            std.lv = TRUE, std.ov = TRUE, auto.fix.single = FALSE, se = "none")

# note it requires smaller penalties than other applications
reg.out2 = cv_regsem(model = semFit, pars_pen = "loadings",
                    mult.start = TRUE, multi.iter = 10,
                    n.lambda = 100, type = "lasso", jump = 10^-5, lambda.start = 0.001)
reg.out2
plot(reg.out2) # note that the solution jumps around -- make sure best fit makes sense
## End(Not run)