get.factor.model: Factor model estimation via Principal Component Analysis

In hdbinseg: Change-Point Analysis of High-Dimensional Time Series via Binary Segmentation

Factor model estimation via Principal Component Analysis

Description

Estimates the components of the factor structure for an input time series, such as loadings and factors, as well as estimating the number of factors.

Usage

get.factor.model(
  z,
  r.max = NULL,
  ic = c("ah", "bn")[1],
  ic.op = 2,
  normalisation = TRUE
)

Arguments

z	input data matrix, with each row representing the component time series
r.max	maximum allowed number of factors
ic	estimator for the factor number; if(ic=='ah') eigenvalue ratio estimator of Ahn and Horenstein (2013) is used, if(ic=='bn') information criterion estimator of Bai and Ng (2002) is used
ic.op	type of the estimator specified by ic
normalisation	if(normalisation==TRUE) rows of z are normalised prior to factor analysis

Value

a list containing

eigvec	eigenvectors of z
eigval	eigenvalues of z
norm.x	row-wise normalised z if(normalisation==TRUE)
r.hat	estimated number of factors
r.max	maximum number of factors used
ic.eval	vector containing information criterion evaluated at r = 0, 1, ..., r.max
mean	row-wise means of z
sd	row-wise standard deviations of z
ic	input ic
ic.op	input ic.op

References

S. C. Ahn and A. R. Horenstein (2013) Eigenvalue ratio test for the number of factors. Econometrica, vol. 81, pp. 1203–1227. J. Bai and S. Ng (2002) Determining the number of factors in approximate factor models. Econometrica, vol. 70, pp. 191–221.

hdbinseg documentation built on Aug. 17, 2023, 5:06 p.m.