nfac.est: Estimate the number of latent factors for exploratory factor...

In abnormally-distributed/cvreg: Cross Validation and Robust Estimation Utilities

Description

This implements the estimator of the number of latent variables for exploratory factor analysis derived by Onatski (2010) from random matrix theory. The number of factors bounded from above using Lederman's boundary, which is the maximum number of factors possible to derive from a given data set.

Usage

nfac.est(Y, max.fac = NULL, max.iter = 10, scale = TRUE)

Arguments

Y	a matrix or data frame of numeric covariates
max.fac	maximum number of factors (if larger than the Lederman boundary this will be ignored)
max.iter	maximum number of iterations. defaults to 10.
scale	should the data be scaled first? defaults to TRUE.

Value

a list containing a numeric value and a vector of the eigenvalue differences.

References

Onatski, A. (2010). Determining the Number of Factors from Empirical Distribution of Eigenvalues. Review of Economics and Statistics, 92(4), 1004–1016. doi:10.1162/rest_a_00043

abnormally-distributed/cvreg documentation built on May 3, 2020, 3:45 p.m.