logPrior

Logarithm of the prior distribution on the number of change-points.

Assume that a temporal process is composed of contiguous segments with differing slopes and replicated noise-corrupted time series measurements are observed. The unknown mean of the data generating process is modelled as a piecewise linear function of time with an unknown number of change-points. The package infers the joint posterior distribution of the number and position of change-points as well as the unknown mean parameters per time-series by MCMC sampling. A-priori, the proposed model uses an overfitting number of mean parameters but, conditionally on a set of change-points, only a subset of them influences the likelihood. An exponentially decreasing prior distribution on the number of change-points gives rise to a posterior distribution concentrating on sparse representations of the underlying sequence, but also available is the Poisson distribution. See Papastamoulis et al (2019) <doi:10.1515/ijb-2018-0052> for a detailed presentation of the method.

Panagiotis Papastamoulis

beast

Bayesian Estimation of Change-Points in the Slope of
Multivariate Time-Series

logPrior function

<dl> <dt>cutPoints</dt>
<dd>change-points.</dd> <dt>nTime</dt>
<dd>number of time-points.</dd> <dt>startPoint</dt>
<dd>optional integer, at least equal to 2.</dd></dl>

Arguments

Author

Log-prior. — logPrior

<dl>

 <dt>cutPoints</dt>
<dd>change-points.</dd>

 <dt>nTime</dt>
<dd>number of time-points.</dd>

 <dt>startPoint</dt>
<dd>optional integer, at least equal to 2.</dd>

</dl>

logPrior: Log-prior.

Description

Usage

Value

Arguments

Author