ks.cp3o_delta {ecp} | R Documentation |
An algorithm for multiple change point analysis that uses dynamic programming and pruning. The Kolmogorov-Smirnov statistic is used as the goodness-of-fit measure.
ks.cp3o_delta(Z, K=1, minsize=30, verbose=FALSE)
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. This is also the window size used to calculate between-segment distances. |
verbose |
A flag indicating if status updates should be printed. |
Segmentations are found through the use of dynamic programming and pruning. Between-segment distances are calculated only using points within a window of the segmentation point.
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. |
Wenyu Zhang
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.
Kifer D., Ben-David S., Gehrke J. (2004). Detecting change in data streams. International Conference on Very Large Data Bases.
set.seed(400)
x = matrix(c(rnorm(100),rnorm(100,3),rnorm(100,0,2)))
y = ks.cp3o_delta(Z=x, K=7, minsize=30, verbose=FALSE)
#View estimated change point locations
y$estimates