select.fanc: Select a Model from a "fanc" Object

In fanc: Penalized Likelihood Factor Analysis via Nonconvex Penalty

Select a Model from a "fanc" Object

Description

Selects the value of rho that minimizes a model-selection criterion, optionally at a specified value of gamma.

Usage

select(x, criterion=c("BIC","AIC","CAIC","EBIC"), 
	gamma,  scores=FALSE, df.method="active")

Arguments

x	Fitted "fanc" model object.
criterion	The criterion by which to select the tuning parameter rho. One of "AIC", "BIC", "CAIC", or "EBIC". Default is "BIC".
gamma	The value of gamma.
scores	Logical flag for outputting factor scores. The original data matrix must have been supplied to fanc. The default is FALSE.
df.method	Two types of degrees of freedom are supported. If "active", the degrees of freedom are the number of nonzero parameters. If "reparametrization", the degrees of freedom of the MC+ are reparametrized based on the degrees of freedom of the lasso.

Value

loadings	factor loadings
uniquenesses	unique variances
Phi	Factor-correlation matrix, when factor correlations were estimated.
scores	Factor scores, when scores=TRUE.
df	Degrees of freedom under the selected df.method.
AIC, BIC, CAIC, EBIC	The value of the selected criterion.
goodness.of.fit	Values of GFI, AGFI, CFI, RMSEA and SRMR, when available.
rho	The selected value of rho.
gamma	The selected value of gamma.

Author(s)

Kei Hirose
[e-mail]

References

Hirose, K. and Yamamoto, M. (2015). Sparse estimation via nonconcave penalized likelihood in a factor analysis model,
Statistics and Computing, 25, 633–648.

See Also

fanc and plot.fanc objects.

fanc documentation built on July 26, 2026, 9:06 a.m.