Bayesian analysis of a change-point in exponential families with applications

Lee, Chung-Bow

doi:10.1016/s0167-9473(98)00009-7

Cited by 26 publications

(19 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Therefore, the change point is estimated at 40. Several researches have been done with different approaches such that Bayesian analysis [28], likelihood ratio [29] and stochastic approximation [30] on the same dataset. In the comparison from the literature, a similar result has been carried out that the change point at 40 or year 1890 which is also met by Barry and Hartigan's approach.…”

Section: Resultsmentioning

confidence: 99%

Bayesian analysis of change point problems for time series data

Lee¹,

Ong²

2016

AIP Conference Proceedings

View full text Add to dashboard Cite

Abstract. Alterations or sudden changes within a sequence of temporal observations always create disturbance to data analysis. The maneuver to detect this alterations or changes in any temporal data may allow researchers to identify the aberration in every block of segments. The Bayesian method proposed by Barry and Hartigan has greatly fitted the analysis of change point problems through product partition model. We study Bayesian analysis for change point problem with Markov sampling computation on British coal mine accident. The result provides accurate change point and posterior means estimation.

show abstract

Section: Resultsmentioning

confidence: 99%

Bayesian analysis of change point problems for time series data

Lee¹,

Ong²

2016

AIP Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…Raftery and Akman (1986) found a change-point, 1890. Lee (1998) used Bayesian analysis and found that the change-point was at 1890, the same change-point as Siegmund (1988) estimated with the likelihood ratio approach. The means and standard deviations of the two identified segments are (3.0976, 0.9014) and (1.5938, 1.0303) respectively.…”

Section: British Coal-mine Accident Data For the Poisson Change-pointmentioning

confidence: 98%

Bayesian Multiple Change-Point Estimation and Segmentation

Kim¹,

Cheon²

2013

Communications for Statistical Applications and Methods

View full text Add to dashboard Cite

This study presents a Bayesian multiple change-point detection approach to segment and classify the observations that no longer come from an initial population after a certain time. Inferences are based on the multiple change-points in a sequence of random variables where the probability distribution changes. Bayesian multiple change-point estimation is classifies each observation into a segment. We use a truncated Poisson distribution for the number of change-points and conjugate prior for the exponential family distributions. The Bayesian method can lead the unsupervised classification of discrete, continuous variables and multivariate vectors based on latent class models; therefore, the solution for change-points corresponds to the stochastic partitions of observed data. We demonstrate segmentation with real data.

show abstract

“…is based on the posterior distribution of the change-point k [3,10]. The advantage of this procedure is that asymptotics is not required [6], unlike in non-Bayesian classical procedures [4,5].…”

Section: Statistical Model and Test Proceduresmentioning

confidence: 99%

“…West and Ogden [13] used the HUS data for the estimation problem of change-points in a sequence of Poisson random variables approached by allowing the change to range over a continuous time interval 0; T , and derived a Bayesian-based interval estimator. Lee [10] proposed a Bayesian test for an exponential random family based on the Type II maximum likelihood (ML-II) approach. The objectives of these works are dierent from our work.…”

Section: Introductionmentioning

confidence: 99%