Repeatedly splits the series at the point maximising the energy
statistic of Matteson and James (2014), eqs. (5)-(7), stopping when
a within-segment permutation test no longer rejects at level
sig. Detects changes in the full distribution rather than
only the mean, and needs no parametric model.
Usage
morie_e_div(
x,
sig = 0.05,
R = 199L,
alpha = 1,
min_size = 2L,
max_cp = NULL,
seed = 20260809
)Arguments
- x
Numeric vector, or a matrix with one observation per row.
- sig
Significance level of the permutation stopping rule.
- R
Number of permutations per test.
- alpha
Energy exponent, in
(0, 2).- min_size
Minimum segment length.
- max_cp
Optional cap on the number of change points.
- seed
Seed of the counter-based stream shared with the Python arm, so the permutation test is reproducible across languages.
Value
A list with changepoints (in discovery order),
changepoints_sorted, p_values, q_stats,
n_changepoints, estimate, n, method.
References
Matteson, D. S. and James, N. A. (2014). A nonparametric approach for multiple change point analysis of multivariate data. JASA, 109(505), 334-345.
Examples
V <- c(1, 2, 3, 4, 5, 6, 7, 8)
morie_e_div(V)
#> $changepoints
#> numeric(0)
#>
#> $changepoints_sorted
#> numeric(0)
#>
#> $p_values
#> [1] 0.055
#>
#> $q_stats
#> [1] 9.333333
#>
#> $n_changepoints
#> [1] 0
#>
#> $estimate
#> numeric(0)
#>
#> $n
#> [1] 8
#>
#> $method
#> [1] "E-divisive (Matteson-James 2014)"
#>