Calculate regime-switching auto-correlation
acf_rs(x, label, lag_max, demean = TRUE)
x
: A univariate numeric time series.label
: A factor of regime labels.lag_max
: Maximum lag at which to calculate the acf.demean
: Logical. Should the covariances be about the sample means?Mean auto-correlations for each group in label
.
set.seed(123) x <- rnorm(100) label <- sample(1:2, 100, replace = TRUE) acf_rs(x, label = factor(label), lag_max = 3)
Useful links