Model Introduced SPRT for Structural Change Detection of Time Series (I) -- Formulation --

Koyama, Yoshihide; Hattori, Tetsuo; Kawano, Hideaki

doi:10.2991/jrnal.2014.1.1.11

Cited by 4 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order for the image processing to have such arrangement capability, we would like to introduce some related processing methods of Signal Processing in Kansei Engineering, Time Series Analysis, and Advection Diffusion in Fluid Dynamics. [11][12][13][14][15]…”

Section: Resultsmentioning

confidence: 99%

Automated Processing of Multiple-Brightness Peak Histogram Image Using Curvature and Variance Estimation

Kawakami¹,

Hattori²,

Imai³

et al. 2016

JRNAL

Self Cite

View full text Add to dashboard Cite

Previously, we have illustrated that the Histogram Matching based on Gaussian Distribution (HMGD) is an effective automated image processing method for obtaining a better feeling impression image. However, the simple HMGD works only for the image whose histogram has just one peak. For the image whose histogram has multiplebrightness peak, it does not work as in the case of single peak histogram image. In this paper, we propose the improved method for multiple-brightness peak (HMGD-MBP). This method can not only detect multiple peaks but also estimate the variance of Gaussian distribution at each detected peak in the image histogram, using curvature computation. This paper also presents the effectiveness of the proposed method by showing the experimental results.

show abstract

Section: Resultsmentioning

confidence: 99%

Automated Processing of Multiple-Brightness Peak Histogram Image Using Curvature and Variance Estimation

Kawakami¹,

Hattori²,

Imai³

et al. 2016

JRNAL

Self Cite

View full text Add to dashboard Cite

show abstract

“…The CPD method based on retrospective analysis requires not only the observation data before the moment to be measured but also the subsequent data of the moment. The detection mechanism is generally based on the probability distribution test of the time series, and the more typical methods are: Bayesian test [32], [33], [34], likelihood ratio test [35], [36], [37] and cumulative sum test [38], [39]. The original Bayesian method was offline [33], [34].…”

Section: Literature Reviewmentioning

confidence: 99%

“…Adams et al first proposed an online Bayesian approach under the hypothesis that the time series data are independently and identically distributed and gave a calculation method to predict the probability distribution of the data at the next moment based on the data value at the current moment [32]. Koyama et al, for time series with a Poisson distribution, give a new strategy for calculating the likelihood ratio of the series, and a change point decision is achieved by hypothesis testing [35]. Compared with the CPD method based on real-time analysis, the method based on retrospective analysis has better robustness and more accurate detection results.…”

Section: Literature Reviewmentioning

confidence: 99%

Detecting Pattern Changes in Individual Travel Behavior Based on a Bayesian Method

Chen,

Li,

Sun

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This paper focuses on the long-overlooked phenomenon that individual travel patterns are not always stable over the long term and may change due to seasonal changes, moving, and work schedule changes. Unlike previous studies that identified sudden peak points, this paper treats travel pattern change as a change point detection problem in a time series and defines change as "sudden, substantial, and continuous". Considering the complexity of travel behavior, this paper measures changes in travel patterns in three dimensions: time, space, and frequency, and establishes a Bayesian change point detection model. A nine-month period of private car GPS data from Aichi, Japan, is used for an example analysis. The results show that the Bayesian approach can effectively identify travel pattern changes. Compared with the traditional GLR, the proposed method in this paper has higher recognition accuracy with lower model complexity. Meanwhile, the experimental results show that individual travel patterns may change in only one dimension or in multiple dimensions at the same time. Based on this, the correlation analysis of travel patterns in the temporal and spatial dimensions is carried out, and it is verified that there is a certain positive correlation between the two. The Bayesian change-point detection model proposed is robust and generally applicable to other fields besides travel patterns.INDEX TERMS Travel pattern, change point detection, Bayesian method, travel behavior analysis

show abstract

Model Introduced SPRT for Structural Change Detection of Time Series (II)

Koyama¹,

Hattori²,

Takeda³

et al. 2014

JRNAL

Self Cite

View full text Add to dashboard Cite

In this paper, using the notion of a binary Channel Matrix as well known in Information Theory, we present an equivalent relation between the SPRT (Sequential Probability Ratio Test) and Bayes' Updating. Moreover, we show the relationship between the SPRT and NSPR (New Sequential Probability Ratio) where a Hidden Markov Model with Poisson process is introduced as structural change model. And we also provide the change point detection performance of SPRT and NSPR by experimental results.

show abstract

Model Introduced SPRT for Structural Change Detection of Time Series (I) -- Formulation --

Cited by 4 publications

References 9 publications

Automated Processing of Multiple-Brightness Peak Histogram Image Using Curvature and Variance Estimation

Automated Processing of Multiple-Brightness Peak Histogram Image Using Curvature and Variance Estimation

Detecting Pattern Changes in Individual Travel Behavior Based on a Bayesian Method

Model Introduced SPRT for Structural Change Detection of Time Series (II)

Contact Info

Product

Resources

About