GCCfactor (version 1.0.1)

BS_local_comp: Get a bootstrap confidence interval for the global component

Description

This function employs a bootstrap procedure to obtain confidence intervals for the local component for the \(j\)-th individual in block \(i\) at time \(t\). See Lin and Shin (2023) for details.

Usage

BS_local_comp(object, i, j, t, BB = 599, alpha = 0.05)

Value

A matrix containing the upper and lower band.

Arguments

object

An S3 object of class 'multi_result' created by multilevel().

i

An integer indicating the \(i\)-th block.

j

An integer indicating the \(j\)-th individual in the \(i\)-th block.

t

An integer specifying the time point at which the CI is constructed.

BB

An integer indicating the number of bootstrap repetition. 599 by default.

alpha

The significance level, a single numeric between 0 and 1. 0.05 by default.

References

Lin, R. and Shin, Y., 2022. Generalised Canonical Correlation Estimation of the Multilevel Factor Model. Available at SSRN 4295429.

Examples

# \donttest{
panel <- UKhouse # load the data
est_multi <- multilevel(panel, ic = "BIC3", standarise = TRUE, r_max = 5,
                           depvar_header = "dlPrice", i_header = "Region",
                           j_header = "LPA_Type", t_header = "Date")
bs_fcomp_111 <- BS_local_comp(est_multi, i = 1, j = 1, t = 1)
# }