pelt.norm.meanvar.cost: Calculate Normal likelihood of data segment, assuming...

In benpickering/smop: Subset Multivariate Optimal Partitioning

Description

Calculates the Normal likelihood of multiple segments of data, as defined by the vector of possible start-points (tau) and a single common end-point (R). This calculation sets both the mean and variance equal to their maximum likelihood estimates.

Calculates the Normal likelihood of multiple segments of data, as defined by the vector of possible start-points (tau) and a single common end-point (R). This calculation sets both the mean and variance equal to their maximum likelihood estimates.

Usage

pelt.norm.meanvar.cost(tau, R, sumx)

pelt.norm.meanvar.cost(tau, R, sumx)

Arguments

tau	A vector of locations indicating the possible beginnings of a segment.
R	A single location representing the end of the possible segment
sumx	The summary statistics for the entire time series being analysed.
tau	A vector of locations indicating the possible beginnings of a segment.
R	A single location representing the end of the possible segment
sumx	The summary statistics for the entire time series being analysed.

Value

A vector containing the cost of the different possible segments defined by tau and R.

A vector containing the cost of the different possible segments defined by tau and R.

Examples

data = rnorm(100, 10, 3)
sumx = pelt.norm.sum(data)
tau = c(30,40,50) # start-points of possible segments
R = 70 # end-point of possible segments
pelt.norm.meanvar.cost(tau,R,sumx) # costs for each segment
data = rnorm(100, 10, 3)
sumx = pelt.norm.sum(data)
tau = c(30,40,50) # start-points of possible segments
R = 70 # end-point of possible segments
pelt.norm.meanvar.cost(tau,R,sumx) # costs for each segment

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