bcp: An R Package for Performing a Bayesian Analysis of Change Point Problems

By Chandra Erdman, John W. Emerson. 2007

384 citations recorded by the source corpus.

Barry and Hartigan (1993) propose a Bayesian analysis for change point problems. We provide a brief summary of selected work on change point problems, both preceding and following Barry and Hartigan. We outline Barry and Hartigan's approach and offer a new R package, bcp (Erdman and Emerson 2007), implementing their analysis. We discuss two frequentist alternatives to the Bayesian analysis, the recursive circular binary segmentation algorithm (Olshen and Venkatraman 2004) and the dynamic programming algorithm of (Bai and Perron 2003). We illustrate the application of bcp with economic and microarray data from the literature.

Subfields: Artificial Intelligence, Molecular Biology, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/eda5b1e85ba73c77cc685cf94940a2b992f9b1d0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Molecular Biology

In Statistics and Probability

Citations in the corpus

changepoint: An R Package for Changepoint Analysis (2014, background): “At the time of writing several such packages exist, including those which provide a single test statistic e.g., sde (Iacus, 2009), bcp (Erdman and Emerson, 2007) and/or are designed for a speciﬁc (typically genomic) application e.g., cumSeg (Muggeo, 2011), DNAcopy (Seshan and Olshen, 2008).”
A review of change point detection methods (2018, methodology): “Computations are usually performed using the Viterbi algorithm or by random sampling methods (such as Monte Carlo Markov Chain, MCMC [46, 48]) if prior distributions on the parameters are provided.”
Change Points and Temporal Dependence in Reconstructions of Annual Temperature: Did Europe Experience a Little Ice Age? (2014, methodology): “We apply the modification of Escanciano and Lobato (2009b) where correlations are divided by sample autocovariances of the squared series to provide robustness against heteroskedasticity, and p is chosen by a data-dependent procedure where a penalty term is subtracted that switches between Akaike an”
ecp: An R Package for Nonparametric Multiple Change Point Analysis of Multivariate Data (2013, methodology): “Unlike the changepoint and cpm packages, the bcp package (Erdman and Emerson, 2007) is designed to perform Bayesian single change point analysis of univariate time series.”