Machine learning and in particular deep learning algorithms are the emerging approaches to data analysis. These techniques have transformed traditional data mining-based analysis radically into a learning-based model in which existing data sets along with their cluster labels (i.e., train set) are learned to build a supervised learning model and predict the cluster labels of unseen data (i.e., test set). In particular, deep learning techniques are capable of capturing and learning hidden features in a given data sets and thus building a more accurate prediction model for clustering and labeling problem. However, the major problem is that time series data are often unlabeled and thus supervised learning-based deep learning algorithms cannot be directly adapted to solve the clustering problems for these special and complex types of data sets. To address this problem, this paper introduces a two-stage method for clustering time series data. First, a novel technique is introduced to utilize the characteristics (e.g., volatility) of given time series data in order to create labels and thus be able to transform the problem from unsupervised learning into supervised learning. Second, an autoencoder-based deep learning model is built to learn and model both known and hidden features of time series data along with their created labels to predict the labels of unseen time series data. The paper reports a case study in which financial and stock time series data of selected 70 stock indices are clustered into distinct groups using the introduced two-stage procedure. The results show that the proposed procedure is capable of achieving 87.5\% accuracy in clustering and predicting the labels for unseen time series data.
Time series analysis has achieved great success in diverse applications such as network security, environmental monitoring, and medical informatics. Learning similarities among different time series is a crucial problem since it serves as the foundation for downstream analysis such as clustering and anomaly detection. It often remains unclear what kind of distance metric is suitable for similarity learning due to the complex temporal dynamics of the time series generated from event-triggered sensing, which is common in diverse applications, including automated driving, interactive healthcare, and smart home automation. The overarching goal of this paper is to develop an unsupervised learning framework that is capable of learning task-aware similarities among unlabeled event-triggered time series. From the machine learning vantage point, the proposed framework harnesses the power of both hierarchical multi-scale sequence autoencoders and Gaussian Mixture Model (GMM) to effectively learn the low-dimensional representations from the time series. Finally, the obtained similarity measure can be easily visualized for explaining. The proposed framework aspires to offer a stepping stone that gives rise to a systematic approach to model and learn similarities among a multitude of event-triggered time series. Through extensive qualitative and quantitative experiments, it is revealed that the proposed method outperforms state-of-the-art methods considerably.
This chapter discusses correlation analysis of stationary multivariate Gaussian time series in the spectral or Fourier domain. The goal is to identify the hub time series, i.e., those that are highly correlated with a specified number of other time series. We show that Fourier components of the time series at different frequencies are asymptotically statistically independent. This property permits independent correlation analysis at each frequency, alleviating the computational and statistical challenges of high-dimensional time series. To detect correlation hubs at each frequency, an existing correlation screening method is extended to the complex numbers to accommodate complex-valued Fourier components. We characterize the number of hub discoveries at specified correlation and degree thresholds in the regime of increasing dimension and fixed sample size. The theory specifies appropriate thresholds to apply to sample correlation matrices to detect hubs and also allows statistical significance to be attributed to hub discoveries. Numerical results illustrate the accuracy of the theory and the usefulness of the proposed spectral framework.
We apply methods from randomized numerical linear algebra (RandNLA) to develop improved algorithms for the analysis of large-scale time series data. We first develop a new fast algorithm to estimate the leverage scores of an autoregressive (AR) model in big data regimes. We show that the accuracy of approximations lies within $(1+\mathcal{O}(\varepsilon))$ of the true leverage scores with high probability. These theoretical results are subsequently exploited to develop an efficient algorithm, called LSAR, for fitting an appropriate AR model to big time series data. Our proposed algorithm is guaranteed, with high probability, to find the maximum likelihood estimates of the parameters of the underlying true AR model and has a worst case running time that significantly improves those of the state-of-the-art alternatives in big data regimes. Empirical results on large-scale synthetic as well as real data highly support the theoretical results and reveal the efficacy of this new approach. To the best of our knowledge, this paper is the first attempt to establish a nexus between RandNLA and big time series data analysis.
Time series have attracted widespread attention in many fields today. Based on the analysis of complex networks and visibility graph theory, a new time series forecasting method is proposed. In time series analysis, visibility graph theory transforms time series data into a network model. In the network model, the node similarity index is an important factor. On the basis of directly using the node prediction method with the largest similarity, the node similarity index is used as the weight coefficient to optimize the prediction algorithm. Compared with the single-point sampling node prediction algorithm, the multi-point sampling prediction algorithm can provide more accurate prediction values when the data set is sufficient. According to results of experiments on four real-world representative datasets, the method has more accurate forecasting ability and can provide more accurate forecasts in the field of time series and actual scenes.
Generative Adversarial Net (GAN) has been proven to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions can be both categorical and continuous variables containing different kinds of auxiliary information. Our simulation studies show that CGAN is able to learn different kinds of normal and heavy tail distributions, as well as dependent structures of different time series and it can further generate conditional predictive distributions consistent with the training data distributions. We also provide an in-depth discussion on the rationale of GAN and the neural network as hierarchical splines to draw a clear connection with the existing statistical method for distribution generation. In practice, CGAN has a wide range of applications in the market risk and counterparty risk analysis: it can be applied to learn the historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES) and predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate CGAN is able to outperform the Historic Simulation, a popular method in market risk analysis for the calculation of VaR. CGAN can also be applied in the economic time series modeling and forecasting, and an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN is given at the end of the paper.
Hate speech is ubiquitous on the Web. Recently, the offline causes that contribute to online hate speech have received increasing attention. A recurring question is whether the occurrence of extreme events offline systematically triggers bursts of hate speech online, indicated by peaks in the volume of hateful social media posts. Formally, this question translates into measuring the association between a sparse event series and a time series. We propose a novel statistical methodology to measure, test and visualize the systematic association between rare events and peaks in a time series. In contrast to previous methods for causal inference or independence tests on time series, our approach focuses only on the timing of events and peaks, and no other distributional characteristics. We follow the framework of event coincidence analysis (ECA) that was originally developed to correlate point processes. We formulate a discrete-time variant of ECA and derive all required distributions to enable analyses of peaks in time series, with a special focus on serial dependencies and peaks over multiple thresholds. The analysis gives rise to a novel visualization of the association via quantile-trigger rate plots. We demonstrate the utility of our approach by analyzing whether Islamist terrorist attacks in Western Europe and North America systematically trigger bursts of hate speech and counter-hate speech on Twitter.
This paper surveys state-of-the-art methods and models dedicated to time series analysis and modeling, with the final aim of prediction. This review aims to offer a structured and comprehensive view of the full process flow, and encompasses time series decomposition, stationary tests, modeling and forecasting. Besides, to meet didactic purposes, a unified presentation has been adopted throughout this survey, to present decomposition frameworks on the one hand and linear and nonlinear time series models on the other hand. First, we decrypt the relationships between stationarity and linearity, and further examine the main classes of methods used to test for weak stationarity. Next, the main frameworks for time series decomposition are presented in a unified way: depending on the time series, a more or less complex decomposition scheme seeks to obtain nonstationary effects (the deterministic components) and a remaining stochastic component. An appropriate modeling of the latter is a critical step to guarantee prediction accuracy. We then present three popular linear models, together with two more flexible variants of the latter. A step further in model complexity, and still in a unified way, we present five major nonlinear models used for time series. Amongst nonlinear models, artificial neural networks hold a place apart as deep learning has recently gained considerable attention. A whole section is therefore dedicated to time series forecasting relying on deep learning approaches. A final section provides a list of R and Python implementations for the methods, models and tests presented throughout this review. In this document, our intention is to bring sufficient in-depth knowledge, while covering a broad range of models and forecasting methods: this compilation spans from well-established conventional approaches to more recent adaptations of deep learning to time series forecasting.
Dynamic Linear Models (DLMs) are commonly employed for time series analysis due to their versatile structure, simple recursive updating, and probabilistic forecasting. However, the options for count time series are limited: Gaussian DLMs require continuous data, while Poisson-based alternatives often lack sufficient modeling flexibility. We introduce a novel methodology for count time series by warping a Gaussian DLM. The warping function has two components: a transformation operator that provides distributional flexibility and a rounding operator that ensures the correct support for the discrete data-generating process. Importantly, we develop conjugate inference for the warped DLM, which enables analytic and recursive updates for the state space filtering and smoothing distributions. We leverage these results to produce customized and efficient computing strategies for inference and forecasting, including Monte Carlo simulation for offline analysis and an optimal particle filter for online inference. This framework unifies and extends a variety of discrete time series models and is valid for natural counts, rounded values, and multivariate observations. Simulation studies illustrate the excellent forecasting capabilities of the warped DLM. The proposed approach is applied to a multivariate time series of daily overdose counts and demonstrates both modeling and computational successes.
Domains such as manufacturing and medicine crave for continuous monitoring and analysis of their processes, especially in combination with time series as produced by sensors. Time series data can be exploited to, for example, explain and predict concept drifts during runtime. Generally, a certain data volume is required in order to produce meaningful analysis results. However, reliable data sets are often missing, for example, if event streams and times series data are collected separately, in case of a new process, or if it is too expensive to obtain a sufficient data volume. Additional challenges arise with preparing time series data from multiple event sources, variations in data collection frequency, and concept drift. This paper proposes the GENLOG approach to generate reliable event and time series data that follows the distribution of the underlying input data set. GENLOG employs data resampling and enables the user to select different parts of the log data to orchestrate the training of a recurrent neural network for stream generation. The generated data is sampled back to its original sample rate and is embedded into a template representing the log data format it originated from. Overall, GENLOG can boost small data sets and consequently the application of online process mining.
Add to Chrome
Add to Firefox