The Development of an Information Criterion for Change-Point Analysis

Change-point analysis is a flexible and computationally tractable tool for the analysis of times series data from systems that transition between discrete states and whose observables are corrupted by noise. The change point algorithm is used to identify the time indices (change points) at which the system transitions between these discrete states. We present a unified information-based approach to testing for the existence of change points. This new approach reconciles two previously disparate approaches to change-point analysis (frequentist and information based) for testing transitions between states. The resulting method is statistically principled, parameter and prior free, and widely applicable to a wide range of change-point problems.

Download Full-text

CHANGE POINT ANALYSIS: A STATISTICAL APPROACH TO DETECT POTENTIAL ABRUPT CHANGE

Jurnal Teknologi ◽

10.11113/jt.v79.10388 ◽

2017 ◽

Vol 79 (5) ◽

Cited By ~ 2

Author(s):

Siti Nur Afiqah Mohd Arif ◽

Mohamad Farhan Mohamad Mohsin ◽

Azuraliza Abu Bakar ◽

Abdul Razak Hamdan ◽

Sharifah Mastura Syed Abdullah

Keyword(s):

Relative Humidity ◽

Change Point ◽

Abrupt Change ◽

Carbon Dioxide Concentration ◽

Change Points ◽

Change Point Analysis ◽

Total Rainfall ◽

Essential Information ◽

Point Analysis ◽

Annual Means

Change-point analysis has proven to be an efficient tool in understanding the essential information contained in meteorological data, such as rainfall, ozone level, and carbon dioxide concentration. In this study, change-point analysis was used to discover potential significant changes in the annual means of total rainfall, temperature and relative humidity from 25 years of Malaysian climate data. Two methods, the CUSUM and bootstrap, were used in the analysis, where the CUSUM was used to analyze the data trends and patterns and bootstrapping was used to calculate the occurrence of change points based on the confidence level. The results of the analysis showed that potential abrupt shifts seem to have taken place in 1999, 2001 and 2002 with respect to the annual means for relative humidity, temperature and total rainfall, respectively. These identified change points will be further analyzed as potential candidates of abrupt change by extending the proposed method in a future study.

Download Full-text

Change-Point Analysis of Tropical Night Occurrences for Five Major Cities in Republic of Korea

Advances in Meteorology ◽

10.1155/2015/801981 ◽

2015 ◽

Vol 2015 ◽

pp. 1-11 ◽

Cited By ~ 1

Author(s):

Myoung-Seok Suh ◽

Chansoo Kim

Keyword(s):

Time Series ◽

Change Point ◽

Time Series Data ◽

Republic Of Korea ◽

Change Point Analysis ◽

Series Data ◽

Daily Minimum Temperature ◽

Rapid Urbanization ◽

Tropical Night ◽

Point Analysis

Bayesian change-point analysis is applied to detect a change-point in the occurrences of tropical night (TN) days in the 50-year time series data for five major cities in Republic of Korea. A TN day is simply defined as a day when the daily minimum temperature is greater than 25∘C. A Bayesian analysis is performed for detecting a change-point at an unknown time point in the TN day frequency time series, which is modeled by an independent Poisson random variable. The results showed that a single change occurred around 1993 for three cities (Seoul, Incheon, and Daegu). However, when we excluded the extraordinary year, 1994, a single change occurred around 1993 only in Seoul and Daegu. The average number of TN days in Seoul and Daegu increased significantly, by more than 150%, after the change-point year. The abrupt increase in TN day frequency in two cities over Republic of Korea around 1993 may be related to the significant decadal change in the East Asian summer monsoon around the mid 1990s and to rapid urbanization.

Download Full-text

Change Point Detection in Terrorism-Related Online Content Using Deep Learning Derived Indicators

Information ◽

10.3390/info12070274 ◽

2021 ◽

Vol 12 (7) ◽

pp. 274

Author(s):

Ourania Theodosiadou ◽

Kyriaki Pantelidou ◽

Nikolaos Bastas ◽

Despoina Chatzakou ◽

Theodora Tsikrika ◽

...

Keyword(s):

Time Series ◽

Change Point ◽

Hate Speech ◽

Change Point Detection ◽

Change Points ◽

Change Point Analysis ◽

Point Analysis ◽

Depth Analysis ◽

Textual Data ◽

Point Detection

Given the increasing occurrence of deviant activities in online platforms, it is of paramount importance to develop methods and tools that allow in-depth analysis and understanding to then develop effective countermeasures. This work proposes a framework towards detecting statistically significant change points in terrorism-related time series, which may indicate the occurrence of events to be paid attention to. These change points may reflect changes in the attitude towards and/or engagement with terrorism-related activities and events, possibly signifying, for instance, an escalation in the radicalization process. In particular, the proposed framework involves: (i) classification of online textual data as terrorism- and hate speech-related, which can be considered as indicators of a potential criminal or terrorist activity; and (ii) change point analysis in the time series generated by these data. The use of change point detection (CPD) algorithms in the produced time series of the aforementioned indicators—either in a univariate or two-dimensional case—can lead to the estimation of statistically significant changes in their structural behavior at certain time locations. To evaluate the proposed framework, we apply it on a publicly available dataset related to jihadist forums. Finally, topic detection on the estimated change points is implemented to further assess its effectiveness.

Download Full-text

Profiles of cognitive change in preclinical Alzheimer’s disease using change-point analysis

10.1101/19009696 ◽

2019 ◽

Cited By ~ 1

Author(s):

Owen A Williams ◽

Yang An ◽

Nicole M Armstrong ◽

Melissa Kitner-Triolo ◽

Luigi Ferrucci ◽

...

Keyword(s):

Alzheimer’S Disease ◽

Alzheimer's Disease ◽

Change Point ◽

Cognitive Change ◽

Change Points ◽

Change Point Analysis ◽

Cognitive Measures ◽

Predictive Tool ◽

Preclinical Ad ◽

Point Analysis

AbstractIntroductionChange-point analyses are increasingly used to identify the temporal stages of accelerated cognitive decline in the preclinical stages of Alzheimer’s Disease (AD). However, statistical comparisons of change-points between specific cognitive measures have not been reported.Methods165 older adults (baseline age range: 61.1-91.2) from the Baltimore Longitudinal Study of Aging developed AD during follow-up. Linear and non-linear mixed models were fit for 11 cognitive measures to determine change-points in rates of decline before AD diagnosis. Bootstrapping was used to compare the timing of change-points across cognitive measures.ResultsChange-points followed by accelerated decline ranged from 15.5 years (Card Rotations) to 1.9 years (Trail-Making A) before AD diagnosis. Accelerated decline in Card Rotations occurred significantly earlier than all other measures, including learning and memory measures.DiscussionResults suggest that visuospatial ability, as assessed by Card Rotations, may have the greatest utility as an early predictive tool in identifying preclinical AD.

Download Full-text

Global Incidence and Mortality Rate of COVID-19; Special Focus on Iran, Italy and China

Archives of Iranian Medicine ◽

10.34172/aim.2020.42 ◽

2020 ◽

Vol 23 (7) ◽

pp. 455-461

Author(s):

Samira Chaibakhsh ◽

Asma Pourhoseingholi ◽

Mohsen Vahedi

Keyword(s):

Mortality Rate ◽

Change Point ◽

World Health ◽

Change Points ◽

Change Point Analysis ◽

Special Focus ◽

Point Analysis ◽

Incidence And Mortality ◽

The World ◽

Health Organization

Background: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a new coronavirus, was diagnosed in China in December 2019. Around the globe, a total of 71429 were infected up to February 17, 2020, with 98.9% of cases in China. On March 11, 2020, the World Health Organization (WHO) characterized the COVID-19 as ‘pandemic’. Rapid positive worldwide incidence was the motivation behind this study to investigate the incidence and mortality globally. Methods: We used the data published by the WHO until March 9, 2020. Non-parametric tests and change point analysis were used for inferences. Results: Change point analysis for Iran and China and the world excluding China for the first 20 days revealed around 78, 195 and 2 further new cases per day, respectively. Italy had a big jump in incidence on the 36th day. Similarly, a sharp rise of positive cases was reported for the world on the 35th day. China successfully controlled the ascending reports of incidence on the 23rd day. Mortality in China and the world were almost similar for the first 20 days. There was an ascending incidence trend with two change points in Italy (30th and 36th days) and one change point in Iran on the 17th day. Mortality in the world jumped remarkably after day 42 with an estimation of almost more than 25 deaths per day. Conclusion: The incidence of COVID-19 varied by regions; however, after March 11, it became ‘pandemic’. It was observed that after about 6 days with an emergence of sharp increase in incidences, there would be a mutation in mortality rate. On the other hand, the importance of ‘on-time’ quarantine programs in controlling this virus was confirmed.

Download Full-text

Transdimensional change-point modeling as a tool to investigate uncertainty in applied geophysical inference: An example using borehole geophysical logs

Geophysics ◽

10.1190/geo2012-0384.1 ◽

2013 ◽

Vol 78 (3) ◽

pp. WB89-WB99 ◽

Cited By ~ 8

Author(s):

Anya M. Reading ◽

Kerry Gallagher

Keyword(s):

Change Point ◽

Probability Distributions ◽

Change Points ◽

Change Point Analysis ◽

Data Series ◽

Model Parameters ◽

Point Analysis ◽

Abrupt Changes ◽

Geophysical Logs ◽

Borehole Log

Recently developed methods for inferring abrupt changes in data series enable such change points in time or space to be identified, and also allow us to estimate noise levels of the observed data. The inferred probability distributions of these parameters provide insights into the capacity of the observed data to constrain the geophysical analysis and hence the magnitudes, and likely sources, of uncertainty. We carry out a change-point analysis of sections of four borehole geophysical logs (density, neutron absorption, sonic interval time, and electrical resistivity) using transdimensional Bayesian Markov chain Monte Carlo to sample a model parameter space. The output is an ensemble of values which approximate the posterior distribution of model parameters. We compare the modeled change points, borehole log parameters, and the variance of the noise distribution of each log with the observed lithology classes down the borehole to make an appraisal of the uncertainty characteristics inherent in the data. Our two examples, one with well-defined lithology changes and one with more subtle contrasts, show quantitatively the nature of the lithology contrasts for which the geophysical borehole log data will produce a detectable response in terms of inferred change points. We highlight the different components of variation in the observed data: due to the geologic process (dominant lithology changes) that we hope to be able to infer, geologic noise due to variability within each lithology, and analytical noise due to the measurement process. This inference process will be a practical addition to the analytical tool box for borehole and other geophysical data series. It reveals the level of uncertainties in the relationships between the data and the observed lithologies and would be of great use in planning and interpreting the results of subsequent routine processing.

Download Full-text

Change Point Enhanced Anomaly Detection for IoT Time Series Data

Water ◽

10.3390/w13121633 ◽

2021 ◽

Vol 13 (12) ◽

pp. 1633

Author(s):

Elena-Simona Apostol ◽

Ciprian-Octavian Truică ◽

Florin Pop ◽

Christian Esposito

Keyword(s):

Time Series ◽

Anomaly Detection ◽

Change Point ◽

Time Series Data ◽

Multivariate Time Series ◽

Change Point Detection ◽

Change Points ◽

Series Data ◽

Prediction And Forecasting ◽

Point Detection

Due to the exponential growth of the Internet of Things networks and the massive amount of time series data collected from these networks, it is essential to apply efficient methods for Big Data analysis in order to extract meaningful information and statistics. Anomaly detection is an important part of time series analysis, improving the quality of further analysis, such as prediction and forecasting. Thus, detecting sudden change points with normal behavior and using them to discriminate between abnormal behavior, i.e., outliers, is a crucial step used to minimize the false positive rate and to build accurate machine learning models for prediction and forecasting. In this paper, we propose a rule-based decision system that enhances anomaly detection in multivariate time series using change point detection. Our architecture uses a pipeline that automatically manages to detect real anomalies and remove the false positives introduced by change points. We employ both traditional and deep learning unsupervised algorithms, in total, five anomaly detection and five change point detection algorithms. Additionally, we propose a new confidence metric based on the support for a time series point to be an anomaly and the support for the same point to be a change point. In our experiments, we use a large real-world dataset containing multivariate time series about water consumption collected from smart meters. As an evaluation metric, we use Mean Absolute Error (MAE). The low MAE values show that the algorithms accurately determine anomalies and change points. The experimental results strengthen our assumption that anomaly detection can be improved by determining and removing change points as well as validates the correctness of our proposed rules in real-world scenarios. Furthermore, the proposed rule-based decision support systems enable users to make informed decisions regarding the status of the water distribution network and perform effectively predictive and proactive maintenance.

Download Full-text

Detecting multiple generalized change-points by isolating single ones

Metrika ◽

10.1007/s00184-021-00821-6 ◽

2021 ◽

Author(s):

Andreas Anastasiou ◽

Piotr Fryzlewicz

Keyword(s):

Model Selection ◽

Linear Trend ◽

Information Criterion ◽

Change Points ◽

Isolation Technique ◽

New Approach ◽

Piecewise Constant ◽

Practical Performance ◽

R Packages ◽

The Times

AbstractWe introduce a new approach, called Isolate-Detect (ID), for the consistent estimation of the number and location of multiple generalized change-points in noisy data sequences. Examples of signal changes that ID can deal with are changes in the mean of a piecewise-constant signal and changes, continuous or not, in the linear trend. The number of change-points can increase with the sample size. Our method is based on an isolation technique, which prevents the consideration of intervals that contain more than one change-point. This isolation enhances ID’s accuracy as it allows for detection in the presence of frequent changes of possibly small magnitudes. In ID, model selection is carried out via thresholding, or an information criterion, or SDLL, or a hybrid involving the former two. The hybrid model selection leads to a general method with very good practical performance and minimal parameter choice. In the scenarios tested, ID is at least as accurate as the state-of-the-art methods; most of the times it outperforms them. ID is implemented in the R packages IDetect and breakfast, available from CRAN.

Download Full-text