norm.meanvar.seglen: Multivariate Cost Calculation: Normally distributed...

In benpickering/smop: Subset Multivariate Optimal Partitioning

Description

Calculates the cost between a sequence of multivariate observations using the provided cumulative sums of data, with the additional penalty of the segment length (log(n)). The observations are assumed to be Normally distributed. The mean and variance parameters are set to their maximum likelihood estimates. Used for detecting changes in both mean and variance.

Usage

norm.meanvar.seglen(n, x, x2, p, length.checklist)

Arguments

n	A matrix containing the lengths between the changepoints in current changepoint vector being considered and the changepoints in the check-list of previous changepoint vectors.
x	A p x length.checklist matrix containing the cumulative sums of the observations between the current changepoint vector and each corresponding changepoint vector on the check-list.
x2	The same as x, except containing the cumulative sums of the squared observations.
p	The number of variables in the series.
length.checklist	The number of possible previous changepoint vectors on the check-list being considered.

Details

An internal function, not designed for use by the end-user.

Value

A vector of length length.checklist containing the multivariate costs for each changepoint vector on the check-list being considered.

benpickering/smop documentation built on Sept. 4, 2020, 1:45 a.m.