Search

Time series measurements from sensing units (e.g., UWB ranging circuits) always suffer from uncertainties like noises, outliers, dropouts, and/or nonspecific anomalies. In order to extract the true information with high precision from the original corrupted measurements, the signal-model-based signal pre-processing units embedded in sensing circuits are usually employed. However, for a general signal to observe, its signal model cannot be obtained so that the signal-model-based signal processing methods are not applicable. In this article, the time-variant local autocorrelated polynomial (TVLAP) model in the state space is proposed to model the dynamics of a non-stationary stochastic process (i.e., a signal or a time series), through which the model-based signal processing methods could be utilized to denoise, to correct the outliers/dropouts, and/or to identify anomalies contained in the measurements. Besides, the presented method can also be used in change point detection for a time series. [ABSTRACT FROM AUTHOR]

We develop a new method to find the number of volatility regimes in a nonstationary financial time series by applying unsupervised learning to its volatility structure. We use change point detection to partition a time series into locally stationary segments and then compute a distance matrix between segment distributions. The segments are clustered into a learned number of discrete volatility regimes via an optimization routine. Using this framework, we determine the volatility clustering structure for financial indices, large-cap equities, exchange-traded funds and currency pairs. Our method overcomes the rigid assumptions necessary to implement many parametric regime-switching models while effectively distilling a time series into several characteristic behaviours. Our results provide a significant simplification of these time series and a strong descriptive analysis of prior behaviours of volatility. Finally, we create and validate a dynamic trading strategy that learns the optimal match between the current distribution of a time series and its past regimes, thereby making online risk-avoidance decisions at present. [ABSTRACT FROM AUTHOR]

This paper proposes a new method of detecting the gradual changes of time series when the changes in time series are mixed with seasonality. The key of the proposed method is to express the desired time-varying feature while removing the unwanted time-varying feature of seasonal effects through two-stage procedures. Asymptotic properties of the proposed methods are studied, and simulation results are presented. In addition, models with multiple changes have been studied. Furthermore, to demonstrate the usefulness of the proposed method, real data analysis with the number of Korean traveling to Japan is presented. [ABSTRACT FROM AUTHOR]

Continual development necessitates innovation. One must discover seeds of innovation and then concentrate resources on these seeds. To do so, one must recognize technology trends and then adopt and execute appropriate innovation strategies. This study used advanced change point detection method to investigate leading indicators that represent changes in technology trends. We examine patents, papers, and newspaper articles in Japan and US for 55 technologies. Results suggest that patents can be more appropriate as leading indicators than either papers or newspapers. This result can contribute to appropriate innovation strategies for planning and updating, and can provide tools that are useful to decision-makers. [ABSTRACT FROM AUTHOR]

Understanding the state transition of a process from the time series data obtained from the process is important from the viewpoint of both analyzing and controlling the process. In particular, it is important to clarify a turning point of the state transition, that is, the point of change in the process and to find the cause of the state transition by observing and analyzing the data from the process. This paper considers the method of detecting several change points in a process based on the likelihood theory and information criterion when a series of data from the process follows the Poisson process. Then, the method of finding any process state fluctuation and points of change is called “the process state tracking method”. The validity and applicability of the process state tracking method introduced in this paper is confirmed through some numerical applications. [ABSTRACT FROM AUTHOR]

In this work, we introduce the R package fabisearch , available on the Comprehensive R Archive Network (CRAN), which implements an original change point detection method for multivariate high-dimensional time series data and a new interactive, 3-dimensional, brain-specific network visualization capability in a flexible, stand-alone function. Change point detection is a commonly used technique in time series analysis, capturing the dynamic nature in which many real-world processes function. With the ever increasing troves of multivariate high-dimensional time series data, especially in neuroimaging and finance, there is a clear need for scalable and data-driven change point detection methods. Currently, change point detection methods for multivariate high-dimensional data are scarce, with even less available in high-level, easily accessible software packages. fabisearch , which implements the factorized binary search (FaBiSearch) methodology, is a novel statistical method for detecting change points in the network structure of multivariate high-dimensional time series which employs non-negative matrix factorization (NMF), an unsupervised dimension reduction and clustering technique. We utilize a new binary search algorithm to efficiently identify multiple change points and provide a new method for network estimation for data between change points. We show the functionality of the package and the practicality of the method by applying it to a neuroimaging and a finance data set. We also introduce an interactive, 3-dimensional, brain-specific network visualization capability in a flexible, stand-alone function. This function can be conveniently used with any node coordinate atlas, and nodes can be color coded according to community membership (if applicable). The output is a network laid over a cortical surface, which can be rotated in 3-dimensional space. [ABSTRACT FROM AUTHOR]

This article presents a novel approach for detecting a change point in the service rate of an M / M / 1 / m queue. The change point is identified by detecting a change in the probability distribution of a stochastic process. To achieve this, the article proposes a likelihood function that is constructed based on the observed number of customers left in the system at departures. Bayesian estimators are then derived using this likelihood function. The effectiveness of the proposed methods is evaluated through extensive Monte Carlo simulations. Overall, the results demonstrate the effectiveness of the proposed approach for detecting change points in service rates. • Change-point estimation is a useful tool for analyzing performance in queues. • Bayesian methods are suited for estimating change points in finite Markovian queues. • Accurate estimates can be obtained from queue lengths at departure times. • Bayesian methods outperform maximum likelihood estimation. [ABSTRACT FROM AUTHOR]

By informing the onset of the degradation process, health status evaluation serves as a significant preliminary step for reliable remaining useful life (RUL) estimation of complex equipment. However, existing works rely on a priori knowledge to roughly identify the starting time of degradation, termed the change point, which overlooks individual degradation characteristics of devices working in variable operating conditions. Consequently, reliable RUL estimation for devices under variable operating conditions is challenging as different devices exhibit heterogeneous and frequently changing degradation dynamics. This paper proposes a novel temporal dynamics learning-based model for detecting change points of individual devices, even under variable operating conditions, and utilises the learnt change points to improve the RUL estimation accuracy. During offline model development, the multivariate sensor data are decomposed to learn fused temporal correlation features that are generalisable and representative of normal operation dynamics across multiple operating conditions. Monitoring statistics and control limit thresholds for normal behaviour are dynamically constructed from these learnt temporal features for the unsupervised detection of device-level change points. The detected change points then inform the degradation data labelling for training a long short-term memory (LSTM)-based RUL estimation model. During online monitoring, the temporal correlation dynamics of a query device is monitored for breach of the control limit derived in offline training. If a change point is detected, the device's RUL is estimated with the well-trained offline model for early preventive action. Using C-MAPSS turbofan engines as the case study, the proposed method improved the accuracy by 5.6% and 7.5% for two scenarios with six operating conditions, when compared to existing LSTM-based RUL estimation models that do not consider heterogeneous change points. • Temporal learning for change point detection under complex, varying conditions. • Unsupervised method does not need domain expertise or train data labelling. • In-depth temporal dynamics is investigated within a degradation process. • A health status-cognizant RUL estimation model is designed with change points. [ABSTRACT FROM AUTHOR]

Abstract Background Biochemical networks are often described through static or time-averaged measurements of the component macromolecules. Temporal variation in these components plays an important role in both describing the dynamical nature of the network as well as providing insights into causal mechanisms. Few methods exist, specifically for systems with many variables, for analyzing time series data to identify distinct temporal regimes and the corresponding time-varying causal networks and mechanisms. Results In this study, we use well-constructed temporal transcriptional measurements in a mammalian cell during a cell cycle, to identify dynamical networks and mechanisms describing the cell cycle. The methods we have used and developed in part deal with Granger causality, Vector Autoregression, Estimation Stability with Cross Validation and a nonparametric change point detection algorithm that enable estimating temporally evolving directed networks that provide a comprehensive picture of the crosstalk among different molecular components. We applied our approach to RNA-seq time-course data spanning nearly two cell cycles from Mouse Embryonic Fibroblast (MEF) primary cells. The change-point detection algorithm is able to extract precise information on the duration and timing of cell cycle phases. Using Least Absolute Shrinkage and Selection Operator (LASSO) and Estimation Stability with Cross Validation (ES-CV), we were able to, without any prior biological knowledge, extract information on the phase-specific causal interaction of cell cycle genes, as well as temporal interdependencies of biological mechanisms through a complete cell cycle. Conclusions The temporal dependence of cellular components we provide in our model goes beyond what is known in the literature. Furthermore, our inference of dynamic interplay of multiple intracellular mechanisms and their temporal dependence on one another can be used to predict time-varying cellular responses, and provide insight on the design of precise experiments for modulating the regulation of the cell cycle.

Under the background of intensifying global change, the frequent occurrence of agrometeorological disasters has an adverse impact on the social economy. Low-temperature weather in spring is one of the main agrometeorological disasters in Xinjiang. Studying the spatial and temporal characteristics of low temperatures in spring in Xinjiang is of great significance. However, research on the spatiotemporal variation of spring frost in arid areas is currently lacking, which limits our understanding of the occurrence and development mechanism of typical disastrous weather events in arid areas. Therefore, on the basis of the daily meteorological data of 40 meteorological stations in the Xinjiang Province of China from 1971 to 2020, we studied the spatiotemporal heterogeneity of spring frost in Xinjiang through trend analysis, the Mann–Kendall test, Kriging interpolation, and wavelet analysis. Results showed that the frequency of spring frost had the spatial trend from Northern Xinjiang to Southern Xinjiang. The occurrence frequency of spring frost in the entire and northern region of Xinjiang had an upward trend, whereas that in Southern Xinjiang showed a downward trend. Severe and moderate spring frost events mainly occurred in Xinjiang. The duration of spring frost had the characteristics of Northern Xinjiang to Southern Xinjiang. The spring frost in Northern Xinjiang mainly occurred in April, and that of Southern Xinjiang was in March. Obvious 15-, 10-, and 20-year oscillation cycles were observed in the occurrence frequency of spring frost in the entire, southern, and northern regions of Xinjiang, respectively. This study can provide a useful reference for the prediction and research corresponding to the occurrence mechanism of spring frost in arid areas.

Homogenization is an important and crucial step to improve the usage of observational data for climate analysis. This work is motivated by the analysis of long series of GNSS Integrated Water Vapour (IWV) data, which have not yet been used in this context. This paper proposes a novel segmentation method called segfunc that integrates a periodic bias and a heterogeneous, monthly varying, variance. The method consists in estimating first the variance using a robust estimator and then estimating the segmentation and periodic bias iteratively. This strategy allows for the use of the dynamic programming algorithm, which is the most efficient exact algorithm to estimate the change point positions. The performance of the method is assessed through numerical simulation experiments. It is implemented in the R package GNSSseg, which is available on the CRAN. This paper presents the application of the method to a real data set from a global network of 120 GNSS stations. A hit rate of 32% is achieved with respect to available metadata. The final segmentation is made in a semi-automatic way, where the change points detected by three different penalty criteria are manually selected. In this case, the hit rate reaches 60% with respect to the metadata.

The stage‐discharge rating curve is subject at many hydrometric stations to sudden changes (shifts) typically caused by morphogenic floods. We propose an original method for estimating shift times using the stage‐discharge observations, also known as gaugings. This method is based on a recursive segmentation procedure that accounts for both gaugings and rating curve uncertainties through a Bayesian framework. It starts with the estimation of a baseline rating curve using all available gaugings. Then it computes the residuals between the gaugings and this rating curve with uncertainties. It proceeds with the segmentation of the time series of residuals through a multi‐change point Bayesian estimation accounting for residuals uncertainties. Once the first set of shift times is identified, the same procedure is recursively applied to each sub‐period through a "top‐down" approach searching for all effective shifts. The proposed method is illustrated using the Ardèche River at Meyras in France (a typical hydrometric site subject to river bed degradation) and evaluated using several synthetic data sets for which the true shift times are known. The applications confirm the added value of the recursive segmentation compared with a "single‐pass" approach and highlight the importance of properly accounting for uncertainties in the segmented data. The recursive procedure effectively disentangles rating changes from observational and rating curve uncertainties. Plain Language Summary: For many hydrological and hydraulic issues, such as flood forecasting, a reliable river discharge estimate is needed. In general, discharge is derived from the recorded water level (stage) through a stage‐discharge relation (rating curve). This relation is calibrated using direct observations (gaugings). Unfortunately, the rating curve is not only uncertain but it can also be subject to sudden changes or shifts due for example to intense floods that modify the river bed geometry. One solution to identify periods of rating curve stability is to apply a segmentation procedure to the gaugings. We propose in this paper an original recursive segmentation procedure that accounts for both gaugings and rating curve uncertainties. Key Points: We propose a method for detecting rating shifts through the segmentation of residuals between the gaugings and a reference rating curveThe method accounts for observational and rating curve uncertainties and expresses change points in terms of time (rather than position)The method recursively re‐estimates the reference rating curve to improve the detection of small shifts [ABSTRACT FROM AUTHOR]

• Change point detection is effective in identifying policy lag on mobility. • Investigating the length of policy lag and magnitude of impact jointly is crucial. • The National declaration of emergency had immediate effect on mobility. • Stay-at-home and reopening policies had lead effect; public responded earlier. • Providing lessons learned in planning and resource allocation for future outbreaks. The unprecedented challenges caused by the COVID-19 pandemic demand timely action. However, due to the complex nature of policy making, a lag may exist between the time a problem is recognized and the time a policy has its impact on a system. To understand this lag and to expedite decision making, this study proposes a change point detection framework using likelihood ratio, regression structure and a Bayesian change point detection method. The objective is to quantify the time lag effect reflected in transportation systems when authorities take action in response to the COVID-19 pandemic. Using travel patterns as an indicator of policy effectiveness, the length of policy lag and magnitude of policy impacts on the road system, mass transit, and micromobility are investigated through the case studies of New York City (NYC), and Seattle—two U.S. cities significantly affected by COVID-19. The quantitative findings show that the National declaration of emergency had no policy lag while stay-at-home and reopening policies had a lead effect on mobility. The magnitude of impact largely depended on the land use and sociodemographic characteristics of the area, as well as the type of transportation system. [ABSTRACT FROM AUTHOR]

In this paper, a sequential change point detection method is developed to monitor structural change in smoothly clipped absolute deviation (SCAD) penalized quantile regression (SPQR) models. The asymptotic properties of the test statistic are derived from the null and alternative hypotheses. In order to improve the performance of the SPQR method, we propose a post‐SCAD penalized quantile regression estimator (P‐SPQR) for high‐dimensional data. We examined the finite sample properties of the proposed methods via Monte Carlo studies under different scenarios. A real data application is provided to demonstrate the effectiveness of the method. [ABSTRACT FROM AUTHOR]

In time series data analysis, detecting change points on a real-time basis (online) is of great interest in many areas, such as finance, environmental monitoring, and medicine. One promising means to achieve this is the Bayesian online change point detection (BOCPD) algorithm, which has been successfully adopted in particular cases in which the time series of interest has a fixed baseline. However, we have found that the algorithm struggles when the baseline irreversibly shifts from its initial state. This is because with the original BOCPD algorithm, the sensitivity with which a change point can be detected is degraded if the data points are fluctuating at locations relatively far from the original baseline. In this paper, we not only extend the original BOCPD algorithm to be applicable to a time series whose baseline is constantly shifting toward unknown values but also visualize why the proposed extension works. To demonstrate the efficacy of the proposed algorithm compared to the original one, we examine these algorithms on two real-world data sets and six synthetic data sets. [ABSTRACT FROM AUTHOR]

Copyright of Fire Ecology is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The mixture cure model has been widely applied to survival data in which a fraction of the observations never experience the event of interest, despite long-term follow-up. In this paper, we study the Cox proportional hazards mixture cure model where the covariate effects on the distribution of uncured subjects' failure time may jump when a covariate exceeds a change point. The nonparametric maximum likelihood estimation is used to obtain the semiparametric estimates. We employ a two-step computational procedure involving the Expectation-Maximization algorithm to implement the estimation. The consistency, convergence rate and asymptotic distributions of the estimators are carefully established under technical conditions and we show that the change point estimator is n consistency. The m out of n bootstrap and the Louis algorithm are used to obtain the standard errors of the estimated change point and other regression parameter estimates, respectively. We also contribute a test procedure to check the existence of the change point. The finite sample performance of the proposed method is demonstrated via simulation studies and real data examples. [ABSTRACT FROM AUTHOR]

Annotating activities of daily living (ADL) is vital for developing machine learning models for activity recognition. In addition, it is critical for self-reporting purposes such as in assisted living where the users are asked to log their ADLs. However, data annotation becomes extremely challenging in real-world data collection scenarios, where the users have to provide annotations and labels on their own. Methods such as self-reports that rely on users' memory and compliance are prone to human errors and become burdensome since they increase users' cognitive load. In this article, we propose a light yet effective context-aware change point detection algorithm that is implemented and run on a smartwatch for facilitating data annotation for high-level ADLs. The proposed system detects the moments of transition from one to another activity and prompts the users to annotate their data. We leverage freely available Bluetooth lowenergy (BLE) information broadcasted by various devices to detect changes in environmental context. This contextual information is combined with a motion-based change point detection algorithm, which utilizes data from wearable motion sensors, to reduce the false positives and enhance the system's accuracy. Through real-world experiments, we show that the proposed system improves the quality and quantity of labels collected from users by reducing human errors while eliminating users' cognitive load and facilitating the data annotation process. [ABSTRACT FROM AUTHOR]

Predictive Maintenance (PdM) is one of the most important applications of advanced data science in Industry 4.0, aiming to facilitate manufacturing processes. To build PdM models, sufficient data, such as condition monitoring and maintenance data of the industrial application, are required. However, collecting maintenance data is complex and challenging as it requires human involvement and expertise. Due to time constraints, motivating workers to provide comprehensive labeled data is very challenging, and thus maintenance data are mostly incomplete or even completely missing. In addition to these aspects, a lot of condition monitoring data-sets exist, but only very few labeled small maintenance data-sets can be found. Hence, our proposed solution can provide additional labels and offer new research possibilities for these data-sets. To address this challenge, we introduce MEDEP, a novel maintenance event detection framework based on the Pruned Exact Linear Time (PELT) approach, promising a low false-positive (FP) rate and high accuracy results in general. MEDEP could help to automatically detect performed maintenance events from the deviations in the condition monitoring data. A heuristic method is proposed as an extension to the PELT approach consisting of the following two steps: (1) mean threshold for multivariate time series and (2) distribution threshold analysis based on the complexity-invariant metric. We validate and compare MEDEP on the Microsoft Azure Predictive Maintenance data-set and data from a real-world use case in the welding industry. The experimental outcomes of the proposed approach resulted in a superior performance with an FP rate of around 10% on average and high sensitivity and accuracy results.

With the rapid development of wind and photovoltaic power generation, hydro-turbine generator units have to operate in a challenging way, resulting in obvious vibration problems. Because of the significant impact of vibration on safety and economical operation, it is of great significance to study the causal relationship between vibration and other variables. The complexity of the hydro-turbine generator unit makes it difficult to analyze the causality of the mechanism. This paper studied the correlation based on a data-driven method, then transformed the correlation into causality based on the mechanism. In terms of correlation, traditional research only judges whether there is a correlation between all data. When the data with correlation are interfered with by the data without correlation, the traditional methods cannot accurately identify the correlation. A piecewise correlation method based on change point detection was proposed to fill this research gap. The proposed method segmented time series pairs, then analyzed the correlation between subsequences. The causality between power load and vibration of a hydro-turbine generator unit was further analyzed. It indicated that when the power load is less than 200 MW, the causality is weak, and when the power load is greater than 375 MW, the causality is strong. The results show that the causality between vibration and power load is not fixed but piecewise. Furthermore, the piecewise correlation method compensated for the limitation of high variance of the maximum information coefficient.

Electricity is a vital part of our daily life; therefore it is important to avoid irregularities such as the California Electricity Crisis of 2000 and 2001. In this work, we seek to predict anomalies using advanced machine learning algorithms. These algorithms are effective, but computationally expensive, especially if we plan to apply them on hourly electricity market data covering a number of years. To address this challenge, we significantly accelerate the computation of the Gaussian Process (GP) for time series data. In the context of a Change Point Detection (CPD) algorithm, we reduce its computational complexity from O($n^{5}$) to O($n^{2}$). Our efficient algorithm makes it possible to compute the Change Points using the hourly price data from the California Electricity Crisis. By comparing the detected Change Points with known events, we show that the Change Point Detection algorithm is indeed effective in detecting signals preceding major events.

In this article, we investigate the problem of simultaneous change point inference and structure recovery in the context of high dimensional Gaussian graphical models with possible abrupt changes. In particular, motivated by neighborhood selection, we incorporate a threshold variable and an unknown threshold parameter into a joint sparse regression model which combines p ℓ1-regularized node-wise regression problems together. The change point estimator and the corresponding estimated coefficients of precision matrices are obtained together. Based on that, a classier is introduced to distinguish whether a change point exists. To recover the graphical structure correctly, a data-driven thresholding procedure is proposed. In theory, under some sparsity conditions and regularity assumptions, our method can correctly choose a homogeneous or heterogeneous model with high accuracy. Furthermore, in the latter case with a change point, we establish estimation consistency of the change point estimator, by allowing the number of nodes being much larger than the sample size. Moreover, it is shown that, in terms of structure recovery of Gaussian graphical models, the proposed thresholding procedure achieves model selection consistency and controls the number of false positives. The validity of our proposed method is justied via extensive numerical studies. Finally, we apply our proposed method to the S&P 500 dataset to show its empirical usefulness. [ABSTRACT FROM AUTHOR]

We consider an optimal control problem where a Brownian motion with drift is sequentially observed and the sign of the drift coefficient changes at jump times of a symmetric two-state Markov process. The Markov process itself is not observable, and the problem consists of finding a {−1, 1}-valued process that tracks the unobservable process as closely as possible. We present an explicit construction of such a process. [ABSTRACT FROM AUTHOR]

Although the significant role of social networks in communications between individuals has attracted researchers' attention to the social networks, only few authors investigated social network monitoring in their studies. Most of the existing studies in this context suffer from the following three main drawbacks: (1) using the case-based network attributes such as person experiences and departments instead of the main attributes such as network density and centrality attributes, (2) monitoring the social attributes separately with the assumption that they are independent of each other and (3) ignoring detection of real time of change in the network. To overcome the above-mentioned disadvantages, this research develops a statistical method for monitoring the connections among actors in the social networks with the four most important network attributes consisting of (1) network density, (2) degree centrality, (3) betweenness centrality and (4) closeness centrality. To this end, a multivariate exponentially weighted moving average (MEWMA) control chart is used for simultaneous monitoring of these four correlated attributes. Furthermore, since the control chart usually does not alert a signal in the exact time of change due to type II error, this study presents a change point detection method to reduce cost and time required for diagnosing the control chart signal. Eventually, the efficiency of the proposed approach in comparison with the existing methods is evaluated through a simulation procedure. The results indicate that the suggested method has better performance than the univariate approach in detecting change point. [ABSTRACT FROM AUTHOR]

In this article we study the theoretical properties of the simultaneous multiscale change point estimator (SMUCE) in piecewise‐constant signal models with dependent error processes. Empirical studies suggest that in this case the change point estimate is inconsistent, but it is not known if alternatives suggested in the literature for correlated data are consistent. We propose a modification of SMUCE scaling the basic statistic by the long run variance of the error process, which is estimated by a difference‐type variance estimator calculated from local means from different blocks. For this modification we prove model consistency for physical‐dependent error processes and illustrate the finite sample performance by means of a simulation study. [ABSTRACT FROM AUTHOR]

We congratulate the author for this interesting paper which introduces a novel method for the data segmentation problem that works well in a classical change point setting as well as in a frequent jump situation. Most notably, the paper introduces a new model selection step based on finding the 'steepest drop to low levels' (SDLL). Since the new model selection requires a complete (or at least relatively deep) solution path ordering the change point candidates according to some measure of importance, a new recursive variant of the Wild Binary Segmentation (Fryzlewicz in Ann Stat 42:2243–2281, 2014, WBS) named WBS2, has been proposed for candidate generation. [ABSTRACT FROM AUTHOR]

In this paper, we consider sequential online prediction (SOP) for streaming data in the presence of outliers and change points. We propose an INstant TEmporal structure Learning (INTEL) algorithm to address this problem. Our INTEL algorithm is developed based on a full consideration of the duality between online prediction and anomaly detection. We first employ a mixture of weighted Gaussian process models (WGPs) to cover the expected possible temporal structures of the data. Then, based on the rich modeling capacity of this WGP mixture, we develop an efficient technique to instantly learn (capture) the temporal structure of the data that follows a regime shift. This instant learning is achieved only by adjusting one hyper-parameter of the mixture model. A weighted generalization of the product of experts (POE) model is used for fusing predictions yielded from multiple GP models. An outlier is declared once a real observation seriously deviates from the fused prediction. If a certain number of outliers are consecutively declared, then a change point is declared. Extensive experiments are performed using a diverse of real datasets. Results show that the proposed algorithm is significantly better than benchmark methods for SOP in the presence of outliers and change points. [ABSTRACT FROM AUTHOR]

The main objective of this research was to investigate the impact of the construction of Karkheh Dam in 2001 (referred to as the intervention time), on groundwater quality. The time series of total dissolved solids (TDS) and other water quality data including potassium (K+), sodium (Na+), magnesium (Mg2+), calcium (Ca2+), bicarbonate (HCO3−), sulfate (SO42−), and chloride (Cl−) for the period between 1996 and 2012 were analyzed. The magnitude of the trend by Sen’s slope estimator for HCO3−, SO42−, and TDS was 0.005, − 0.02 and − 3.04, where a decline expected for SO42− and TDS, whereas for HCO3−, an increase was expected. According to the Pettitt’s test, the mean of TDS decreased from 2306.9 mg/l between 1996 and 2002 to 797.2 mg/l between 2002 and 2012. During this time, the standard deviation of TDS declined from 2187.1 to 132.0 mg/l. The results of change point detection by the Pruned Exact Linear Time (PELT) algorithm were consistent with that of Pettitt’s test providing confirmation that a change point in Ca2+, Mg2+, SO42−, and TDS time series data occurred in 2002.The findings from intervention analysis using the Bayesian structural time series (BSTS) technique showed that TDS concentration during the post-intervention period had an average value of 1127 mg/l compared with 1972 mg/l, before the dam construction. The time series of TDS demonstrated a decrease of about 43% following the intervention time. [ABSTRACT FROM AUTHOR]

Change point detection is essential to understand the time-evolving structure of dynamic networks. Recent research shows that a latent semantic indexing (LSI)-based algorithm effectively detects the change points of a dynamic network. The LSI-based method involves a singular value decomposition (SVD) on the data matrix. In a dynamic scenario, recomputing the SVD of a large matrix each time new data arrives is prohibitively expensive and impractical. A more efficient approach is to incrementally update the decomposition. However, in the classical incremental SVD (incSVD) algorithm, the information of the newly added columns is not fully considered in updating the right singular space, resulting in an approximation error which cannot be ignored. This paper proposes an enhanced incSVD (EincSVD) algorithm, in which the right singular matrix is calculated in an alternative way. An adaptive EincSVD (AEincSVD) algorithm is also proposed to further reduce the computational complexity. Theoretical analysis proves that the approximation error of the EincSVD is smaller than that of the incSVD. Simulation results demonstrate that the EincSVD and the AEincSVD perform much better than the incSVD on change point detection, and the performance of the EincSVD is comparable to the batch SVD algorithm.

We consider a function estimation method with change point detection using truncated power spline basis and elastic-net-type L1-norm penalty. The L1-norm penalty controls the jump detection and smoothness depending on the value of the parameter. In terms of the proposed estimators, we introduce two computational algorithms for the Lagrangian dual problem (coordinate descent algorithm) and constrained convex optimization problem (an algorithm based on quadratic programming). Subsequently, we investigate the relationship between the two algorithms and compare them. Using both simulation and real data analysis, numerical studies are conducted to validate the performance of the proposed method.

We study the sequential quickest change point detection for systems with multiple possible postchange models. A change point is the time instant at which the distribution of a random process changes. In many practical applications, the prechange model can be easily obtained, yet the postchange distribution is unknown due to the unexpected nature of the change. In this article, we consider the case that the postchange model is from a finite set of possible models. The objective is to minimize the average detection delay (ADD), subject to upper bounds on the probability of false alarm (PFA). Two different quickest change detection algorithms are proposed under Bayesian and non-Bayesian settings. Under the Bayesian setting, the prior probabilities of the change point and prior probabilities of possible postchange models are assumed to be known, yet this information is not available under the non-Bayesian setting. Theoretical analysis is performed to quantify the analytical performance of the proposed algorithms in terms of exact or asymptotic bounds on PFA and ADD. It is shown through theoretical analysis that when PFA is small, both algorithms are asymptotically optimal in terms of ADD minimization for a given PFA upper bound. Numerical results demonstrate that the proposed algorithms outperform existing algorithms in the literature. [ABSTRACT FROM AUTHOR]

We present an efficient score statistic, called the S 3 T statistic, to detect the emergence of a spatially and temporally correlated signal from either fixed-sample or sequential data. The signal may cause a mean shift and/or a change in the covariance structure. The score statistic can capture both the spatial and temporal structures of the change and hence is particularly powerful in detecting weak signals. The score statistic is computationally efficient and statistically powerful. Our main theoretical contribution is accurate analytical approximations to the false alarm rate of the detection procedures, which can be used to calibrate the threshold analytically. Numerical experiments on simulated and real data, as well as a case study of water quality monitoring using sensor networks, demonstrate the good performance of our procedure. [ABSTRACT FROM AUTHOR]

The ability to detect change points is a core skill in system monitoring and prognostics. When data take the form of frequencies, i.e., count data, counting processes such as Poisson processes are extensively used for modeling. However, many existing count-based approaches rely on parametric models or deterministic frameworks, failing to consider complex system uncertainty based on temporal and environmental contexts. Another challenge is analyzing interrelated events simultaneously to detect change points that can be missed by independent analyses. This article presents a Multi-Output Log-Gaussian Cox Process with a Cross-Spectral Mixture kernel (MOLGCP-CSM) as a count-based change point detection algorithm. The proposed model employs MOLGCP to flexibly model time-varying intensities of events over multiple channels with the CSM kernel that can capture either negative or positive correlations, as well as phase differences between stochastic processes. During the monitoring, the proposed approach measures the level of change in real-time by computing a weighted likelihood of observation with respect to the constructed model and determines whether a target system experiences a change point by conducting a statistical test based on extreme value theory. Our method is validated using three types of datasets: synthetic, accelerometer vibration, and gas regulator data. [ABSTRACT FROM AUTHOR]