619 results found
Akansu AN, Malioutov D, Palomar DP, et al., 2016, Introduction to the Issue on Financial Signal Processing and Machin Learning for Electronic Trading, IEEE Journal of Selected Topics in Signal Processing, Vol: 10, Pages: 979-981, ISSN: 1932-4553
The twelve papers in this special issue presents relevant research contributions from the disciplines of finance, mathematics, data science and engineering to facilitate scientific cross-fertilization. It will also serve the signal processing community to be exposed to the state of the art in mathematical finance, financial engineering, financial signal processing an electronic trading, and to foster future research in this emerging area. The main themes of this special issue include using tools from machine learning and signal processing that help to address some of the main problems arising in quantitative finance: modeling risk and correlations of financial instruments and their baskets, returns and liquidity, and problems involving risk-aware resource allocation -namely portfolio optimization. These problems involve tools from convex and discrete optimization, non-parametric statistics, time-series modeling,graph theory and high-dimensional covariance estimation.
Rehman NU, Abbas SZ, Asif A, et al., 2016, Translation invariant multi-scale signal denoising based on goodness-of-fit tests, SIGNAL PROCESSING, Vol: 131, Pages: 220-234, ISSN: 0165-1684
Carrion Garcia A, Chanwimalueang T, Giovanni Calvi G, et al., 2016, Modelling economic stress through financial systemic balance index, Pages: 565-569
© 2016 IEEE. Financial markets undergo cycles oscillating between periods of economic growth followed by periods of recession. This is similar to the principle of sympathovagal balance in humans representing the leverages between two interactive nervous systems: The sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS). Based on the Efficient Market Hypothesis (EMH), we conjecture that there could exist two major groups of people who invest in financial markets: 1. Those who drive financial prices to a state of abnormality, and 2. Those who try to sustain the financial prices. To support this conjecture, we introduce the financial systemic balance index (FSBI) which is used to quantify the evolution of financial stress levels in the market. Results show that periods of higher stress detected by the proposed FSBI index correspond to those with higher levels of determinism and nonlinearity, evaluated via the Delay Vector Variance (DVV) method.
Abstract:Financial markets typically undergo periods of prosperity followed by periods of stagnation, and this undulation makes it challenging to maintain market efficiency. The efficient market hypothesis (EMH) states that there exist differences in structural complexity in security prices between regular and abnormal situations. Yet, despite a clear link between market acceleration (cf. recession in security prices) and stress in physical systems, indices of financial stress still have significant scope for further development. The overarching aim of this work is therefore to determine the characteristics of financial indices related to financial stress, and to establish a robust metric for the extent of such `stress'. This is achieved based on intrinsic multiscale analysis which quantifies the so called complexity-loss hypothesis in the context of financial stress. The multiscale sample entropy and our proposed Assessment of Latent Index of Stress methods have successfully assessed financial stress, and have served as a measure to establish an analogy between transitions from `normal' (relaxed) to `abnormal' (stressed) financial periods with the sympatho-vagal balance in humans. Four major stock indices of the US economy over the past 25 years are considered: (i) Dow Jones Industrial Average, (ii) NASDAQ Composite, (iii) Standard & Poor's 500, and (iv) Russell 2000, together with FTSE 100, CAC 40 and exchange rates. Our findings support the EMH theory and reveal high stress for both the periods of Internet bubble burst and sub-prime mortgage crisis.
Kanna S, Mandic DP, 2016, Steady-State Behavior of General Complex-Valued Diffusion LMS Strategies, IEEE Signal Processing Letters, Vol: 23, Pages: 722-726, ISSN: 1558-2361
A novel methodology to bound the steady-state mean square performance of the diffusion complex least mean square (D-CLMS) and the diffusion widely linear (augmented) CLMS (D-ACLMS) algorithm is proposed. This is achieved by exploiting the almost identical nature of the steady-state filter weights at all nodes. The proposed approach allows for the consideration of the second-order terms in the recursion for the weight error covariance matrix, without compromising the mathematical tractability of the problem. The closed form expressions for the mean square deviation (MSD) and excess mean square error (EMSE) for both the D-CLMS and D-ACLMS allow for the performance of the algorithms to be quantified as a function of the noncircularity of the input data.
Chanwimalueang T, Aufegger L, von Rosenberg W, et al., 2016, Modelling stress in public speaking: evolution of stress levels during conference presentations, International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2016, Publisher: IEEE, Pages: 814-818
The Electrocardiogram (ECG) collected in real-life scenarios is often noisy and contaminated with motion artefacts. This study proposes a new framework to analyse the heart rate variability (HRV) in mobile scenarios by introducing novel R-peak detection and HRV detrending algorithms. The R-peak detection combines matched filtering and Hilbert transform, while detrending the HRV is performed using empirical mode decomposition with novel physically meaningful stopping criteria. Next, four quantitative metrics-sample entropy, LFhrv, HFhrv and LF/HF ratio ??? are used to estimate stress levels in two public speaking events: (i) a presentation in front of an audience and (ii) an interactive poster presentation, both at ICASSP 2015. We show that the proposed framework makes it possible to detect distinctive stress-patterns in the structural complexity of the HRV, thus verifying the complexity-loss hypothesis in physiological research.
Hemakom A, Goverdovsky V, Looney D, et al., 2016, Adaptive-projection intrinsically transformed multivariate empirical mode decomposition in cooperative brain-computer interface applications, Philosophical Transactions of the Royal Society A: Mathematical, Physical & Engineering Sciences, Vol: 374, ISSN: 1364-503X
An extension to multivariate empirical mode decomposition (MEMD), termed adaptive-projection intrinsically transformed MEMD (APIT-MEMD), is proposed to cater for power imbalances and inter-channel correlations in real-world multichannel data. It is shown that the APIT-MEMD exhibits similar or better performance than MEMD for a large number of projection vectors, whereas it outperforms MEMD for the critical case of a small number of projection vectors within the sifting algorithm. We also employ the noise-assisted APIT-MEMD within our proposed intrinsic multiscale analysis framework and illustrate the advantages of such an approach in notoriously noise-dominated cooperative brain–computer interface (BCI) based on the steady-state visual evoked potentials and the P300 responses. Finally, we show that for a joint cognitive BCI task, the proposed intrinsic multiscale analysis framework improves system performance in terms of the information transfer rate.
Tonoyan Y, Looney D, Mandic DP, et al., 2016, Discriminating multiple emotional states from EEG using a data-adaptive, multiscale information-theoretic approach, International Journal of Neural Systems, Vol: 26, ISSN: 1793-6462
A multivariate sample entropy metric of signal complexity is applied to EEG data recorded when subjects were viewing four prior-labeled emotion-inducing video clips from a publically available, validated database. Besides emotion category labels, the video clips also came with arousal scores. Our subjects were also asked to provide their own emotion labels. In total 30 subjects with age range 19–70 years participated in our study. Rather than relying on predefined frequency bands, we estimate multivariate sample entropy over multiple data-driven scales using the multivariate empirical mode decomposition (MEMD) technique and show that in this way we can discriminate between five self-reported emotions (p<0.05p<0.05). These results could not be obtained by analyzing the relation between arousal scores and video clips, signal complexity and arousal scores, and self-reported emotions and traditional power spectral densities and their hemispheric asymmetries in the theta, alpha, beta, and gamma frequency bands. This shows that multivariate, multiscale sample entropy is a promising technique to discriminate multiple emotional states from EEG recordings.
Talebi SP, Kanna S, Xia Y, et al., 2016, A Distributed Quaternion Kalman Filter With Applications to Fly-by-Wire Systems, IEEE International Conference on Digital Signal Processing (DSP), Publisher: IEEE, Pages: 30-34
Xia Y, Wang K, Pei W, et al., 2016, A Least Squares Enhanced Smart DFT Technique for Frequency Estimation of Unbalanced Three-Phase Power Systems, International Joint Conference on Neural Networks (IJCNN), Publisher: IEEE, Pages: 2762-2766, ISSN: 2161-4393
Yavari E, Rahman A, Xu J, et al., 2016, Synchrosqueezing an Effective Method for Analyzing Doppler Radar Physiological Signals, 38th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society (EMBC), Publisher: IEEE, Pages: 263-266, ISSN: 1557-170X
Zhang Z, Li H, Mandic D, 2016, Blind Source Separation and Artefact Cancellation for Single Channel Bioelectrical Signal, 13th IEEE International Conference on Wearable and Implantable Body Sensor Networks (BSN), Publisher: IEEE, Pages: 177-182, ISSN: 2376-8886
Xiang M, Kanna S, Douglas SC, et al., 2016, PERFORMANCE ADVANTAGE OF QUATERNION WIDELY LINEAR ESTIMATION: AN APPROXIMATE UNCORRELATING TRANSFORM APPROACH, IEEE International Conference on Acoustics, Speech, and Signal Processing, Publisher: IEEE, Pages: 4348-4352, ISSN: 1520-6149
Douglas SC, Mandic DP, 2016, STABILITY ANALYSIS OF THE LEAST-MEAN-MAGNITUDE-PHASE ALGORITHM, IEEE International Conference on Acoustics, Speech, and Signal Processing, Publisher: IEEE, Pages: 4935-4939, ISSN: 1520-6149
Enshaeifar S, Took CC, Sanei S, et al., 2016, NOVEL QUATERNION MATRIX FACTORISATIONS, IEEE International Conference on Acoustics, Speech, and Signal Processing, Publisher: IEEE, Pages: 3946-3950, ISSN: 1520-6149
Hemakom A, Goverdovsky V, Aufegger L, et al., 2016, QUANTIFYING COOPERATION IN CHOIR SINGING: RESPIRATORY AND CARDIAC SYNCHRONISATION, IEEE International Conference on Acoustics, Speech, and Signal Processing, Publisher: IEEE, Pages: 719-723, ISSN: 1520-6149
Jaksic V, Mandic DP, Ryan K, et al., 2016, A comprehensive study of the delay vector variance method for quantification of nonlinearity in dynamical systems, Royal Society Open Science, Vol: 3, ISSN: 2054-5703
Although vibration monitoring is a popular method to monitor and assess dynamic structures, quantification of linearity or nonlinearity of the dynamic responses remains a challenging problem. We investigate the delay vector variance (DVV) method in this regard in a comprehensive manner to establish the degree to which a change in signal nonlinearity can be related to system nonlinearity and how a change in system parameters affects the nonlinearity in the dynamic response of the system. A wide range of theoretical situations are considered in this regard using a single degree of freedom (SDOF) system to obtain numerical benchmarks. A number of experiments are then carried out using a physical SDOF model in the laboratory. Finally, a composite wind turbine blade is tested for different excitations and the dynamic responses are measured at a number of points to extend the investigation to continuum structures. The dynamic responses were measured using accelerometers, strain gauges and a Laser Doppler vibrometer. This comprehensive study creates a numerical and experimental benchmark for structurally dynamical systems where output-only information is typically available, especially in the context of DVV. The study also allows for comparative analysis between different systems driven by the similar input.
Mikkelsen KB, Kappel SL, Mandic DP, et al., 2015, EEG Recorded from the Ear: Characterizing the Ear-EEG Method., Frontiers in Neuroscience, Vol: 9, ISSN: 1662-4548
Highlights Auditory middle and late latency responses can be recorded reliably from ear-EEG.For sources close to the ear, ear-EEG has the same signal-to-noise-ratio as scalp.Ear-EEG is an excellent match for power spectrum-based analysis. A method for measuring electroencephalograms (EEG) from the outer ear, so-called ear-EEG, has recently been proposed. The method could potentially enable robust recording of EEG in natural environments. The objective of this study was to substantiate the ear-EEG method by using a larger population of subjects and several paradigms. For rigor, we considered simultaneous scalp and ear-EEG recordings with common reference. More precisely, 32 conventional scalp electrodes and 12 ear electrodes allowed a thorough comparison between conventional and ear electrodes, testing several different placements of references. The paradigms probed auditory onset response, mismatch negativity, auditory steady-state response and alpha power attenuation. By comparing event related potential (ERP) waveforms from the mismatch response paradigm, the signal measured from the ear electrodes was found to reflect the same cortical activity as that from nearby scalp electrodes. It was also found that referencing the ear-EEG electrodes to another within-ear electrode affects the time-domain recorded waveform (relative to scalp recordings), but not the timing of individual components. It was furthermore found that auditory steady-state responses and alpha-band modulation were measured reliably with the ear-EEG modality. Finally, our findings showed that the auditory mismatch response was difficult to monitor with the ear-EEG. We conclude that ear-EEG yields similar performance as conventional EEG for spectrogram-based analysis, similar timing of ERP components, and equal signal strength for sources close to the ear. Ear-EEG can reliably measure activity from regions of the cortex which are located close to the ears, especially in paradigms employing frequency-d
Zhang H, Mandic DP, 2015, Is a complex-valued stepsize advantageous in complex-valued gradient learning algorithms?, IEEE Transactions on Neural Networks and Learning Systems, Vol: 27, Pages: 2730-2735, ISSN: 2162-2388
Complex gradient methods have been widely used in learning theory, and typically aim to optimize real-valued functions of complex variables. The stepsize of complex gradient learning methods (CGLMs) is a positive number, and little is known about how a complex stepsize would affect the learning process. To this end, we undertake a comprehensive analysis of CGLMs with a complex stepsize, including the search space, convergence properties, and the dynamics near critical points. Furthermore, several adaptive stepsizes are derived by extending the Barzilai-Borwein method to the complex domain, in order to show that the complex stepsize is superior to the corresponding real one in approximating the information in the Hessian. A numerical example is presented to support the analysis.
Mandic DP, Kanna S, Constantinides AG, 2015, On the Intrinsic Relationship Between the Least Mean Square and Kalman Filters, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 32, Pages: 117-122, ISSN: 1053-5888
Tobar F, Mandic DP, 2015, Design of Positive-Definite Quaternion Kernels, IEEE SIGNAL PROCESSING LETTERS, Vol: 22, Pages: 2117-2121, ISSN: 1070-9908
Ahrabian A, Mandic DP, 2015, Selective Time-Frequency Reassignment Based on Synchrosqueezing, IEEE SIGNAL PROCESSING LETTERS, Vol: 22, Pages: 2039-2043, ISSN: 1070-9908
Zhou G, Cichocki A, Zhang Y, et al., 2015, Group component analysis for multiblock data: common and individual feature extraction, IEEE Transactions on Neural Networks and Learning Systems, Vol: 27, Pages: 2426-2439, ISSN: 2162-2388
Real-world data are often acquired as a collection of matrices rather than as a single matrix. Such multiblock data are naturally linked and typically share some common features while at the same time exhibiting their own individual features, reflecting the underlying data generation mechanisms. To exploit the linked nature of data, we propose a new framework for common and individual feature extraction (CIFE) which identifies and separates the common and individual features from the multiblock data. Two efficient algorithms termed common orthogonal basis extraction (COBE) are proposed to extract common basis is shared by all data, independent on whether the number of common components is known beforehand. Feature extraction is then performed on the common and individual subspaces separately, by incorporating dimensionality reduction and blind source separation techniques. Comprehensive experimental results on both the synthetic and real-world data demonstrate significant advantages of the proposed CIFE method in comparison with the state-of-the-art.
Tobar F, Djuric PM, Mandic DP, 2015, Unsupervised State-Space Modeling Using Reproducing Kernels, IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol: 63, Pages: 5210-5221, ISSN: 1053-587X
Xu D, Zhang H, Mandic DP, 2015, Convergence analysis of an augmented algorithm for fully complex-valued neural networks, NEURAL NETWORKS, Vol: 69, Pages: 44-50, ISSN: 0893-6080
Miralles R, Carrión A, Looney D, et al., 2015, Characterization of the complexity in short oscillating time series: An application to seismic airgun detonations, Journal of the Acoustical Society of America, Vol: 138, Pages: 1595-1603, ISSN: 0001-4966
© 2015 Acoustical Society of America. Extracting frequency-derived parameters allows for the identification and characterization of acoustic events, such as those obtained in passive acoustic monitoring applications. Situations where it is difficult to achieve the desired frequency resolution to distinguish between similar events occur, for example, in short time oscillating events. One feasible approach to make discrimination among such events is by measuring the complexity or the presence of non-linearities in a time series. Available techniques include the delay vector variance (DVV) and recurrence plot (RP) analysis, which have been used independently for statistical testing, however, the similarities between these two techniques have so far been overlooked. This work suggests a method that combines the DVV method with the recurrence quantification analysis parameters of the RP graphs for the characterization of short oscillating events. In order to establish the confidence intervals, a variant of the pseudo-periodic surrogate algorithm is proposed. This allows one to eliminate the fine details that may indicate the presence of non-linear dynamics, without having to add a large amount of noise, while preserving more efficiently the phase-space shape. The algorithm is verified on both synthetic and real world time series.
Jelfs B, Mandic DP, 2015, A unifying framework for the analysis of proportionate NLMS algorithms, INTERNATIONAL JOURNAL OF ADAPTIVE CONTROL AND SIGNAL PROCESSING, Vol: 29, Pages: 1073-1085, ISSN: 0890-6327
Jaksic V, Mandic DP, Karoumi R, et al., 2015, Estimation of nonlinearities from pseudodynamic and dynamic responses of bridge structures using the Delay Vector Variance method, Physica A - Statistical Mechanics and Its Applications, Vol: 441, Pages: 100-120, ISSN: 0378-4371
Analysis of the variability in the responses of large structural systems and quantification of their linearity or nonlinearity as a potential non-invasive means of structural system assessment from output-only condition remains a challenging problem. In this study, the Delay Vector Variance (DVV) method is used for full scale testing of both pseudo-dynamic and dynamic responses of two bridges, in order to study the degree of nonlinearity of their measured response signals. The DVV detects the presence of determinism and nonlinearity in a time series and is based upon the examination of local predictability of a signal. The pseudo-dynamic data is obtained from a concrete bridge during repair while the dynamic data is obtained from a steel railway bridge traversed by a train. We show that DVV is promising as a marker in establishing the degree to which a change in the signal nonlinearity reflects the change in the real behaviour of a structure. It is also useful in establishing the sensitivity of instruments or sensors deployed to monitor such changes.
Xu D, Jahanchahi C, Took CC, et al., 2015, Enabling quaternion derivatives: the generalized HR calculus, Royal Society Open Science, Vol: 2, ISSN: 2054-5703
Quaternion derivatives exist only for a very restricted class of analytic (regular) functions; however, in many applications, functions of interest are real-valued and hence not analytic, a typical case being the standard real mean square error objective function. The recent HR calculus is a step forward and provides a way to calculate derivatives and gradients of both analytic and non-analytic functions of quaternion variables; however, the HR calculus can become cumbersome in complex optimization problems due to the lack of rigorous product and chain rules, a consequence of the non-commutativity of quaternion algebra. To address this issue, we introduce the generalized HR (GHR) derivatives which employ quaternion rotations in a general orthogonal system and provide the left- and right-hand versions of the quaternion derivative of general functions. The GHR calculus also solves the long-standing problems of product and chain rules, mean-value theorem and Taylor's theorem in the quaternion field. At the core of the proposed GHR calculus is quaternion rotation, which makes it possible to extend the principle to other functional calculi in non-commutative settings. Examples in statistical learning theory and adaptive signal processing support the analysis.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.