e.cp3o: CHANGE POINTS ESTIMATION BY PRUNED OBJECTIVE (VIA E-STATISTIC)

Description

An algorithm for multiple change point analysis that uses dynamic programming and pruning. The E-statistic is used as the goodness-of-fit measure.

Usage

e.cp3o(Z, K=1, minsize=30, alpha=1, verbose=FALSE)

Value

The returned value is a list with the following components.

number: The estimated number of change points.
estimates: The location of the change points estimated by the procedure.
gofM: A vector of goodness of fit values for differing number of change points. The first entry corresponds to when there is only a single change point, the second for when there are two, and so on.
cpLoc: The list of locations of change points estimated by the procedure for different numbers of change points up to K.
time: The total amount to time take to estimate the change point locations.

Arguments

Z: A T x d matrix containing the length T time series with d-dimensional observations.
K: The maximum number of change points.
minsize: The minimum segment size.
alpha: The moment index used for determining the distance between and within segments.
verbose: A flag indicating if status updates should be printed.

Author

Nicholas A. James, Wenyu Zhang

Details

Segmentations are found through the use of dynamic programming and pruning. For long time series, consider using e.cp3o_delta.

References

W. Zhang, N. A. James and D. S. Matteson, "Pruning and Nonparametric Multiple Change Point Detection," 2017 IEEE International Conference on Data Mining Workshops (ICDMW), New Orleans, LA, 2017, pp. 288-295.

Examples

Run this code

set.seed(400)
x1 = matrix(c(rnorm(50),rnorm(50,3)))
y1 = e.cp3o(Z=x1, K=2, minsize=30, alpha=1, verbose=FALSE)
#View estimated change point locations
y1$estimates

Run the code above in your browser using DataLab