Search

Purpose: Causality between stock returns, volatility and traded volume for 10 most liquid stocks from Zagreb Stock Exchange (ZSE) is examined in this paper. Methodology: The paper relies on historical daily data regarding return, standard deviation and turnover for the period from 2015 to 2021. Vector Autoregressive Models (VARs) were estimated for each stock individually. Based on estimated VAR models, Granger-causality tests were performed to estimate causality between trading volume, stock returns and volatility for most liquid stocks from the Croatian stock market. Results: Results strongly confirm that traded volume Granger causes volatility. Return remained irrelevant in terms of predicting traded volume and volatility of stock returns. Conclusion: Causality from return to volatility or causality from volatility to return can be confirmed only in shorter periods. Traded volume causes volatility for the majority of stocks regardless of how volatility was calculated. Causality from volatility to return and causality from volatility to volume are valid for half of the sample and need to be further investigated. [ABSTRACT FROM AUTHOR]

Copyright of Revista de Economia e Sociologia Rural is the property of Sociedade Brasileira de Economia e Sociologia Rural 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.)

In recent years deep learning methods, based on reconstruction errors, have facilitated huge improvements in unsupervised anomaly detection. These methods make the limiting assumption that the greater the distance between an observation and a prediction the lower the likelihood of that observation. In this paper we propose conDENSE, a novel anomaly detection algorithm, which does not use reconstruction errors but rather uses conditional density estimation in masked autoregressive flows. By directly estimating the likelihood of data, our model moves beyond approximating expected behaviour with a single point estimate, as is the case in reconstruction error models. We show how conditioning on a dense representation of the current trajectory, extracted from a variational autoencoder with a gated recurrent unit (GRU VAE), produces a model that is suitable for periodic datasets, while also improving performance on non-periodic datasets. Experiments on 31 time-series, including real-world anomaly detection benchmark datasets and synthetically generated data, show that the model can outperform state-of-the-art deep learning methods. [ABSTRACT FROM AUTHOR]

This paper focuses on development of a three-stage iterative identification algorithm for parameter estimation of a Wiener model with autoregressive moving average (ARMA) noise entering between linear and nonlinear part. The proposed three-stage algorithm is developed based on generalized extended gradient iterative (GEGI) algorithm to increase the convergence rate at a low number of iterations. To increase the convergence rate, a second algorithm is derived based on generalized extended least squares algorithm (GELSI). The proposed 3-stage generalized extended least square iterative (3S-GELSI) algorithm will also decrease the computational burden compared to the single-stage algorithm. Finally, to show that the discussed methods can identify the considered system effectively, simulation results of three examples are provided. The results demonstrate that the 3-stage GEGI (3S-GEGI) algorithm outperforms its basic single-stage counterpart in terms of convergence rate. Moreover, the proposed 3S-GELSI algorithm exhibits even greater superiority in terms of convergence rate compared to other existing methods. Additionally, it will be demonstrated that the computational burden decreases in the 3S-GELSI algorithm, as indicated by the reduced number of total flops. • Identification of output nonlinear-Box Jenkins Wiener models with colored noise. • 3-stage algorithm is developed based on GEGI algorithm. • To increase the convergence rate, an algorithm is derived based on GELSI algorithm. • 3S-GELSI algorithm decrease the computational burden compared to the single-stage algorithm. • The unmeasurable signals and the unknown system parameters are estimated effectively. [ABSTRACT FROM AUTHOR]

The precise diagnostics of different diseases is very important for its treatment. It is particularly important to differentiate the disease on the early stages when the pathological alterations have not yet caused great harm to the whole organism, since this allows using greater number of therapies and increase the recovery probability. One of the methods of early diagnostics is based on examination of human biological liquids (blood, tears, cervical mucus, urine, etc,). A small drop of liquid is drawn on an object-plate and dried out slowly, thus a thin dry film (facies) remains. In the process of fluid crystallization there appear characteristic patterns (markers) in the facies. Each marker is a highly definite sign of some pathology even at an early stage of a disease development. When mass population health examination is carried out, it is necessary to analyze a large number of images This is the reason the problem of algorithm and software development for automated processing of images is rather urgent nowadays. Mathematical models of facies images are needed to formalize the tasks of image processing and synthesis of algorithms. The facies has a round shape, as it is obtained from a drop of liquid. Facies images have a radial-ring structure due to the peculiarities of the drying and crystallization of the contained fractions. These facies features should be taken into account in their imaging models. In this paper, autoregressive models of circular images are used to represent facies images. These models use grids defined on concentric circles or Archimedes' spirals. The radial-ring structure of facies image correlations is naturally taken into account in these models. The paper contains examples of model identification, that is, estimation of the values ​​of its parameters at which the correlation properties of the model are closest to the characteristics of a given facies. It is shown that the classification of the facies type (corresponding to the general condition of the patient) is possible according to these estimates. [ABSTRACT FROM AUTHOR]

Pollution havens create environmental inequality issues. China's recent policy of directing high-pollution firms to migrate out of its capital, Beijing, offers a case of a pollution haven that was mandated, rather than resulting from firms' responses to environmental regulation. More importantly, it leads to a question that has been less discussed in previous literature: does a pollution haven strike back? After the highly polluting enterprises moved out of Beijing to neighboring Hebei province, Hebei has been blamed for the continuing poor air quality in Beijing, based on a perception that the pollutants emitted in Hebei are being blown back to Beijing. This paper investigates the existence of a pollution haven in Hebei and quantifies the contribution of Hebei's emissions to Beijing's air quality. We find that the proportion of output values and the relative concentration of high-pollution industries in Hebei increased when the large-scale out-migration began, while the indicators in Beijing decreased. We also find higher emissions of enterprises in Hebei than their counterparts in Beijing. Together, these findings indicate the existence of a "manmade" pollution haven. Further, based on high-frequency spatial data, we combine an Autoregressive Distributed Lag model and a Spatial Autoregressive Model and decompose the factors that affect Beijing's air quality into local and regional transport of air-borne pollutants. We find a significant effect of the neighboring cities' air pollution on Beijing's air quality. This indicates that the pollution haven indeed struck back. These findings imply the importance of regional policies to control air pollution. [ABSTRACT FROM AUTHOR]

This paper investigates the usefulness of Google search patterns with Artificial Intelligence (AI) techniques for timely influenza-like illness (ILI) forecasting for each of the nine South African provinces. Traditional surveillance methods are limited by delay in reporting. Existing digital disease surveillance studies that employ alternative online data have scarcely explored sub-Saharan African countries. In South Africa, Google search data has only been recently studied for ILI surveillance at the national level. Meanwhile, the differences in socio-economic and technological conditions across provinces call for finer spatial investigation. We perform correlation analysis between Google trends (GT) data for 21 ILI-related terms and real-life ILI surveillance data for each province. Next, we develop models to assess the predictive performance of these GT data for forecasting ILI rates, using time series, machine learning, and deep learning methods. We observe sufficient correlation for only two of the nine provinces: Gauteng and Western Cape. Thus, GT data could only be used to forecast ILI in these two provinces. Interestingly, these two provinces are regarded as the most economically developed. In the other seven provinces, LSTM, a deep learning technique, gives more accurate predictions than a baseline autoregressive model when only past ILI data are used for forecasting future ILI trends. The results reveal that, for provinces for which GT data is sufficiently available, it is not only free and fast, but is an effective predictor on its own as well as when added to past ILI data for forecasting future ILI infection rates. The correlation analysis suggests an association between provincial socio-economic development and the use of digital platforms for disease surveillance. Overall, the study established the need for finer scale ILI forecasting which will inform targeted planning for disease surveillance and interventions. [ABSTRACT FROM AUTHOR]

Machine learning interpretation has a well-established discussion in various areas. Whereas most interpretation is made after the modelling, one may try to obtain the informative by itself model. Mainly such models are represented by time-series models such as AR(auto-regression). For time series, we may also obtain the expression in closed form, which has the advantage of being interpretable. Transferring from single to multiple dimensions increases the search space drastically. Thus, we must reduce it for the initial assumption search. The paper proposes the algorithm of algebraic expression discovery using evolutionary optimization. The initial assumption is made using the Fourier series decomposition. Several examples of the algorithm's work based on known functions and arctic ocean ice concentration data are shown. [ABSTRACT FROM AUTHOR]

In some practical situations, images are set on a circle. For example, images of the facies (thin film) of dried biological fluid, eyes, cut of a tree trunk, etc. Currently, most of the image processing works deal with images defined on rectangular two-dimensional grids or grids of higher dimension. The features of circle images require their consideration in their mathematical models. In this paper, an autoregressive models of homogeneous and inhomogeneous random fields defined on a circle are considered as representations of images with radial or radial-circular structure. In the present paper, autoregressive models of circular images are considered. To represent heterogeneous images with random heterogeneities, «doubly stochastic» models are used in which one or more images control the parameters of the resulting image. Pseudo-gradient algorithms for the modal identification are proposed. The conducted statistical modeling showed that these algorithms give good model identification. [ABSTRACT FROM AUTHOR]

Copyright of Revista de Economia e Sociologia Rural is the property of Sociedade Brasileira de Economia e Sociologia Rural 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.)

Nowadays, image processing problems are becoming increasingly important due to development of the aerospace Earth monitoring systems, medical devices etc. But the most of the image processing works deal with images defined on rectangular two-dimensional grids or grids of higher dimension. In some practical situations, images are set on a circle. The peculiarity of such images requires its consideration in their models. In the early works of the authors, several autoregressive models of such images were proposed. However, these models can generate images with positive covariation functions. Sometimes the impact of the external environment and internal features of the object lead to the presence of harmonic components. For example, facies, eye, virus, sunflower, etc. To represent such images, complex roots are necessary in characteristic equation of autoregressive model. In the present paper, autoregressive models of circular quasiperiodic images are considered. The expressions of their covariance functions are given. The pseudo-gradient algorithms for model identification, and images forecasting and filtering are proposed. [ABSTRACT FROM AUTHOR]

Abstract A computationally efficient algorithm for hinging hyperplane autoregressive exogenous (HHARX) model identification via mixed-integer programming technique is proposed in this paper. The HHARX model is attractive since it accurately approximates a general nonlinear process as a sum of hinge functions and preserves the continuity even in a piecewise affine form. Traditional mixed-integer programming-based method for HHARX model identification can only be applied on small-scale input/output datasets due to its significant computational demands. The contribution of this paper is to develop a sequential optimization approach to build accurate HHARX model more efficiently on a relatively large number of experimental data. Moreover, the proposed framework can handle more difficult and practical cases in piecewise model identification, such as: limited submodel switching, missing output data and specified steady state. Finally, the efficiency and accuracy of the proposed computational scheme are demonstrated through modeling of two simulated examples and a pilot-scale heat exchanger. Highlights • A mixed-integer programming approach is presented to identify HHARX model for nonlinear process. • This method can handle large-scale data set. • The submodel is identified sequentially to achieve high fitness. • The submodel switching frequency, missing output data and specified steady state are considered during the modeling. [ABSTRACT FROM AUTHOR]

Gaussian processes (GP) regression is a powerful probabilistic tool for modeling nonlinear dynamical systems. The downside of the method is its cubic computational complexity with respect to the training data that can be partially reduced using pseudo-inputs. The dynamics can be represented with an autoregressive model, which simplifies the training to that of the static case. When simulating an autoregressive model, the uncertainty is propagated through a nonlinear function and the simulation cannot be evaluated in closed-form. This paper combines the variational methods of GP approximations with a nonlinear autoregressive model with exogenous inputs (NARX) to form variational GP (VGP-NARX) models. We show how VGP-NARX models, on average, better approximate a full GP-NARX model than more commonly used GP-NARX (FITC) model on 10 chaotic time-series. The modeling capabilities of VGP-NARX models are compared with the existing approaches on two benchmarks for modeling nonlinear dynamical systems. The advantage of general-purpose computing on graphics processing units (GPGPU) for Monte Carlo simulation on large validation data sets is addressed. • Variational approximation of GP-NARX models for dynamic system identification. • Comparison of GP-NARX approximations for modeling chaotic time-series. • Comparison of VGP-NARX models with existing approaches on dynamic benchmark data. • General-purpose computing on graphics processing units for GP-NARX simulation. [ABSTRACT FROM AUTHOR]

There is currently no study that focuses solely on the causal relationship between Crop Production Index (CPI) and Permanent Cropland (PCL) in developed and developing countries, especially in Nigeria. However, understanding the causal relationship between CPI and PCL is crucial to both food security and economic growth of any nation. In this paper, we investigated the causal relationship between CPI and PCL time series variables using unrestricted Vector Autoregressive (VAR) modeling techniques. Pre-examination of the CPI and PCL time series data extracted from the repository of World Bank showed that these two series were not only difference stationary series of order one {I(1)} but are also not cointegrated. The results of optimal lag length confirmed that VAR (3) model best fitted the data. The findings showed that Nigeria's crop production index is predictable by Nigeria's permanent cropland and vice versa. [ABSTRACT FROM AUTHOR]

In this study a process of acquiring a trend of significant spectral coefficients of Photonic Doppler Velocimetry (PDV) data is proposed. The novelty of the paper is the design of a methodology which will allow to find a specific curve describing data of aluminium metal plate acceleration by detonation products of brisant high explosive obtained using PDV in time on the basis of frequency response. The paper combine short time Fourier Transform (STFT), time-frequency varying autoregressive process (TFAR) to specify the description of detonation products from both time and frequency perspectives. We also investigate the identification of a curve describing such behavior of frequency response on time in processed spectrogram. [ABSTRACT FROM AUTHOR]

This paper is one of the outcomes of the PicsAR (Physics Augmented Reality) research project that is focusing on the evaluation of students' abstract thinking skills while utilizing augmented reality (AR) in the atomic model. The sequence of the research has emerged into three steps: developing, validating, and the evaluation of the PicsAR. The research utilized an ADDIE model: "Analysis-Design-Development-Implementation-Evaluation". In Spring semester 2019, the authors conducted these steps and resulted in a pocket of PicsAR booklet and AR application based on Android. Then, the trial of application was conducted to 33 students in private high school in Surabaya, Indonesia. Parameters of evaluation included the quality of PicsAR, impact on students abstract thinking skills, and the research outcomes. The results indicated (1) the process of developing AR in atomic model fulfil the criteria of product quality: validity, practicality, and effectiveness, (2) performing of students' abstract thinking skills reached at least 66.67% in the combination of good and very good categories of all reasoning categories, (3) through PicsAR research project resulted in two prior publications and one property right. The atomic model is one of abstract physics concept representative in presenting the use of augmented reality in physics learning; therefore, the recommendation of this research is another abstract physics concepts should address the use of AR as a media for learning. [ABSTRACT FROM AUTHOR]

Universal filtered multi-carrier (UFMC) modula- tion is a very powerful candidate to be employed for future 5G mobile systems. It overcomes the limitations and restric- tions in current modulation techniques employed in 4G mobile systems and supports future applications, such as machine- to-machine (M2M), device-to-device (D2D), and vehicle-to- vehicle (V2V) communications. In this paper, we address the estimation of UFMC fading channels based on the comb-type pilot arrangement in the frequency domain. The basic so- lution is to estimate the fading channel based on the mean square error (MSE) or least square (LS) criteria with adap- tive implementation using least mean square (LMS) or recur- sive least square (RLS) algorithms. However, these adaptive filters seem not to be effective, as they cannot fully exploit fad- ing channel statistics, particularly at high Doppler rates. To take advantage of these statistics, time-variations of the fading channel are modeled by an autoregressive process (AR), and are tracked by an H¥ filter. This, however, requires that AR model parameters be known, which are estimated by solving the Yule-Walker equation (YWE), based on the Bessel autocor- relation function (ACF) of the fading channel with a known Doppler rate. Results of Matlab simulations show that the proposed H¥ filter-based channel estimator is more effective when compared with existing estimators. [ABSTRACT FROM AUTHOR]

Nowadays, image processing problems are becoming increasingly important due to development of the aerospace Earth monitoring systems, radio and sonar systems, medical devices for early diagnosis etc. But the most of the image processing works deal with images defined on rectangular two-dimensional grids or grids of higher dimension. In some practical situations, images are set on a cylinder. For example, images of a section of pipelines, worms, blood vessel, parts during turning, etc. The peculiarity of the domain for specifying such images requires its consideration in their models and processing algorithms. In the present paper, autoregressive models of cylindrical images are considered, expressions of the correlation function depending on the autoregression parameters are given. To represent heterogeneous images with random heterogeneities, «doubly stochastic» models are used in which one or more images control the parameters of the resulting image. The spiral scan of a cylindrical image can be considered as a quasiperiodic process due to the correlation of image rows. Pseudo-gradient algorithms for the modal identification are proposed. The statistical modeling showed that these algorithms give good model identification. [ABSTRACT FROM AUTHOR]

For the severe haze situation in the Beijing-Tianjin-Hebei region, conventional fine particulate matter (PM2.5) concentration prediction methods based on pollutant data face problems such as incomplete data, which may lead to poor prediction performance. Therefore, this paper proposes a method of predicting the PM2.5 concentration based on image analysis technology that combines image data, which can reflect the original weather conditions, with currently popular machine learning methods. First, based on local parameter estimation, autoregressive (AR) model analysis and local estimation of the increase in image blur, we extract features from the weather images using an approach inspired by free energy and a no-reference robust metric model. Next, we compare the coefficient energy and contrast difference of each pixel in the AR model and then use the percentages to calculate the image sharpness to derive the overall mass fraction. Furthermore, the results are compared. The relationship between residual value and PM2.5 concentration is fitted by generalized Gauss distribution (GGD) model. Finally, nonlinear mapping is performed via the wavelet neural network (WNN) method to obtain the PM2.5 concentration. Experimental results obtained on real data show that the proposed method offers an improved prediction accuracy and lower root mean square error (RMSE). [ABSTRACT FROM AUTHOR]

The fractional order grey models have appealed considerable interest of research in recent years due to its high effectiveness and flexibility in time series forecasting. However, the existing fractional order accumulation and difference are computationally complex, which leads to difficulties for theoretical analysis and applications. In this paper, new definitions of fractional accumulation and difference are proposed based on the definition of conformable fractional derivative, which are called the conformable fractional accumulation and difference. Then a novel conformable fractional grey model is proposed based on the conformable fractional accumulation and difference, and Brute Force method is introduced to optimize its fractional order. The feasibility and simplicity of the proposed model and the Brute Force method are shown in the numerical example. The conformable fractional grey model outperforms the existing fractional grey model and the autoregressive model in 1 to 3-step predictions with 21 benchmark data sets, and also outperforms the existing fractional grey model in predicting the natural gas consumption of 11 countries. The results indicate that the proposed conformable fractional grey model is more efficient in longer term prediction and non-smooth time series forecasting than the existing models. • New definitions of Conformable Fractional Accumulation and Difference are proposed. • A novel Conformable Fractional Grey Model is proposed. • Effectiveness of CFGM is validated with benchmark time series datasets. • The CFGM is more sensitive to the new information. • The CFGM has high potential in energy forecasting. [ABSTRACT FROM AUTHOR]

• A novel design of fuzzy-weighted differential evolution based computing paradigm is efficaciously portrayed for the estimation of fractional-order electrically stimulated muscle models parameters, generalized with fractional Hammerstein structure. • The proposed optimization heuristics effectively estimate a static nonlinear block parameters dynamic fractional-order linear block parameters and fractional order of nonlinear Hammerstein controlled autoregressive structures. • The precision of the designed optimization strategy is validated on various noisy environments for the fractional-order electrically stimulated muscle models as well as different type of nonlinearities i.e., polynomial, sigmoidal, and cubic spline in the input signal. • The significance of the presented algorithm is substantiated on convergence analysis, fitness plots, absolute error measurements, distribution analysis, histograms studies and boxplot illustrations, for different variations of nonlinear F-ESM models identification. Fractional order representations model complex systems with less parameters, improved accuracy and enhanced robustness in control systems but the system identification of fractional order systems is highly complicated and challenging task because of substantial nonlinearity of the input, unknown linear/nonlinear parameters of the blocks, and unknown fractional order. In this paper, parameter estimation technique is proposed based on fuzzy-weighted differential evolution algorithm for fractional electrically stimulated muscle models, which is generalization of fractional Hammerstein controlled autoregressive model vital for rehabilitation of spinal cord injury (SCI) patients. The system identification problem of fractional electrically stimulated muscle models (FESMMs) is formulated via mean square error approximation between the true and estimated response of FESMMs. The parameters of FESMMs are identified by employing fuzzy weighted differential evolution algorithm optimization knacks of with polynomial, cubic spline and sigmoidal input nonlinearities on various noise variances in the system dynamics. Comparison of results from actual to calculated responses indicates closeness up to 4 decimal places of accuracy for FESMM with polynomial type nonlinearity, up to 6 decimal places for FESMM with sigmoidal type kernel, and up to 5 decimal for FESMM with cubic spline type nonlinearity for different low and high signal to noise scenarios i.e., σ 2 = 0. 002 2 , 0. 02 2 , 0. 2 2 , which verify the precision, efficacy, stability and reliability of the fuzzy weighted differential evolution algorithm for system identification of FESMMs in rehabilitation scenarios of SCI. [ABSTRACT FROM AUTHOR]

• Dynamic modeling of energy expenditure can capture transient thermoregulation and food intake effect. • LSTM networks promise high accuracy of time-series data prediction and capturing patterns. • Adequate accuracy of dynamic EE prediction should be within 5–10 % of MAPE. • The ensemble of CNN-LSTM and LGBM models can lead to the high accuracy of personalized EE prediction. • Personalized modeling architecture and features selection can lead to capturing transient EE effects in individuals. Dynamic human energy expenditure (EE) consists of energy costs for resting metabolism, food digestion, physical activity, and thermoregulation. Currently, multiple models predict EE mainly concerning physical activity, thus, discarding other factors contributing to the dynamic variation of EE. This paper aimed to demonstrate that (i) a dynamic human body EE prediction requires the time series approach, (ii) personalization of input features and models can outperform the generalized approach. To achieve these objectives, data were collected from 3 sets of experiments with 6 test subjects wearing multiple sensors. The analysis of features' importance showed that the selection of features varies for activity-dominated and non-activity-dominated cases and also varies between individuals. Long-Short Term Memory (LSTM) networks were used to develop personalized models such as a simple LSTM, a convolutional LSTM (CNN-LSTM), and also an ensemble model combining CNN-LSTM with a Gradient Boosting algorithm (LSTM-LGBM). A personalized autoregressive linear model and a generalized approach of the LSTM-LGBM method were also developed to have a base of comparison. The results show that the personalized models provide good prediction accuracy, with the mean absolute percentage error (MAPE) mostly lying in the range of 5–15 %. The CNN-LSTM outperforms a simple LSTM model by 3–5 % in MAPE values, and the ensemble model outperforms the by 5–8 % the simple LSTM. The personalized modeling approach with LSTM has shown the potential to improve the prediction accuracy of dynamic EE and capture the non-activity-related effects such as thermoregulation and postprandial thermogenesis. [ABSTRACT FROM AUTHOR]

In this paper, new algorithms for spectrum sensing in cognitive radio based on higher order cumulants and kurtosis are proposed. The cumulants represent statistical signal processing based on pattern recognition for signals of different structure, and has low implementation complexity. Kurtosis statistics are a well-known technique for testing the Gaussianity feature of the signals. Under the assumption that a detected signal can be modelled according to an autoregressive model, noise variance is estimated from that noisy signal. The simulation results show that spectrum sensing algorithms based on the estimated normalised values of joint higher order cumulants (of fourth and sixth orders) and kurtosis are reliable for a wide range of signal-to-noise ratio environments. In order to improve performances of the spectrum sensing, the combination of these statistics tests into unique one statistic test is proposed. Simulation results have verified improvement of the performances. [ABSTRACT FROM AUTHOR]

Time series of count data is not a widely studied research topic. This paper develops Bayesian forecasting method of counts whose conditional distributions given past observations and explanatory variables follow a Poisson distribution. To determine a forecasted value of an observation-driven integer valued autoregressive model, a series of well thought alghoritm needs to be developed, resulting in the use of Bayesian framework. This effective alghoritm sets is then used for the aforementioned calculation for the modelling of time series count data. To get the demanded results, a particle MCMC algorithm for the autoregressive Poisson regression model is introduced in the equation. Two real-life data sets, monthly demand for medicines type X and type Y (2016–2018) are successfully be analysed. We also illustrate that the Bayesian forecasting is more accurate than the corresponding frequentist's approach. [ABSTRACT FROM AUTHOR]

Copyright of Nova Economia is the property of Nova Economia 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.)

Highlights • A classification scheme to classify focal and non-focal Electroencephalogram (EEG)signals is proposed. • Variational Mode Decomposition and Discrete Wavelet Transform are employed to analyze the EEG signals. • A feature reduction technique based on Neighborhood Component Analysis is used to reduce the number of features. • An ensemble stacking classification approach is adopted and shown that it improves the classification accuracy. • The proposed method can be used as a decision support system in clinical settings to identify focal EEGs. Abstract Classification of focal and non-focal Electroencephalogram (EEG) signals is an important problem especially for the identification of epileptogenic sites in the brain. However, the number of research works reported in the literature is limited and most of them suffer from validation on a limited scale and moderate accuracy. In this paper, focal and non-focal EEG signals are analyzed in variational mode decomposition (VMD) and discrete wavelet transform (DWT) domain and features such as refined composite multiscale dispersion entropy, refined composite multiscale fuzzy entropy, and autoregressive model (AR) coefficients are extracted in VMD, DWT and VMD-DWT domain. Statistical analysis of the features is carried out by one way ANOVA to demonstrate their discriminating ability by means of p -values and visual inspection of the box plots of the features. A feature reduction algorithm based on neighborhood component analysis is used to reduce the model complexity and select the features with the highest discriminating abilities. An ensemble stacking classification approach is adopted to improve the accuracy of classification. The performance of the proposed method is studied using a publicly available benchmark database that contains 3750 pairs of focal and 3750 pairs of non-focal EEG signals. It is shown that the stacking configuration improves the accuracy significantly compared to a standalone classifier. It gives high values of sensitivity (96.1%), Specificity (94.4%), Accuracy (95.2%) and area under curve (0.989). Comparison with various existing methods reveals that the proposed method outperforms the others in accuracy. It may help researchers in developing a computer-aided system to identify epileptogenic sites in the brain. [ABSTRACT FROM AUTHOR]

Abstract Remote heart rate measurement has many powerful applications, such as measuring stress in the workplace, and the analysis of the impact of cognitive tasks on breathing and heart rate variability (HRV). Although many methods are available to measure heart rate remotely from face videos, most of them only work well on stationary subjects under well-controlled conditions, and their performance significantly degrades under subject's motions and illumination variation. We propose a novel algorithm to estimate heart rate. Also, it can differentiate between a photo of a human face and an actual human face meaning that it can detect false signals and skip them. The method obtains ROIs using facial landmarks, then it rectifies illumination based on Normalized Least Mean Square (NLMS) adaptive filter and eliminates non-rigid motions based on standard deviation of fixed length of the signal's segments. The method employs the RADICAL technique to extract independent subcomponents. The heart rate measures for each subcomponent, are estimated by analysis of frequency signal to find the one with the highest magnitude. A two-steps data fusion method is also introduced to combine current and previous measured heart rates to calculate a more accurate result. In this paper, we explore the potential of our algorithm on two self-collected, and DEAP databases. The results of three experiments demonstrate that our algorithm substantially outperforms all previous methods. Moreover, we investigate the behavior of our algorithm under challenging conditions including the subject's motions and illumination variation, which shows that our algorithm can reduce the influences of illumination interference and rigid motions significantly. Also, it indicates that our algorithm can be used for the online environment. Finally, the application of our algorithm in search and rescue scenarios using drones is considered and an experiment is conducted to investigate the algorithm's potential to be embedded in drones. [ABSTRACT FROM AUTHOR]

Highlights • Pattern recognition is utilized to recognize grasping force. • The optimal feature set includes the MAV, RMS and WL. • The six-dimensional vector matrix of the optimal feature set is reduced by LDA. • Quadratic polynomial fitting is used to get continuous values of grasping force. Abstract This paper proposes an EMG recognition system of grasping force on the basis of the pattern recognition, which can classify the surface electromyography (sEMG) signals from 2 electrodes and recognize the grasping force. Ten characteristics in time domain and frequency domain are chosen as the primary features to combine feature sets, to obtain an optimal feature set. The linear discriminant analysis (LDA) is used to reduce the dimension of the features vector to a one-dimensional vector matrix, and pattern recognition to classify and recognize it. In online recognition, to obtain continuous recognition values, the quadratic polynomial fitting is utilized to find the relationship between the one-dimensional vector matrix and grasping forces. [ABSTRACT FROM AUTHOR]

Abstract This paper considers the state estimation of discrete-time Markovian jump nonlinear systems with colored measurement noises obeying a nonlinear autoregressive process of order n , which is motivated by tracking the maneuvering target under electronic countermeasures with high speed sampling or persistent perturbations. In order to remove the measurement noises correlation, the left zero divisor is explored to reconstruct a new measurement equation via difference approach, with the help of applying statistical linear regression to the colored measurement noise model. Then, a novel hypothesis set constituted of all possible values of multi-step Markov jumping parameters is defined and the posterior probability density of the state is derived recursively. By using Gaussian mixtures to approximate the posterior probability densities, an adaptive Gaussian mixture filter for the considered system is proposed, where the Gaussian components with small weights are pruned adaptively through measuring the Alpha (or Beta) divergence for the original and approximated Gaussian mixtures, to achieve a tradeoff between the estimation accuracy and running time. A maneuvering target tracking accompanied by range gate pull-off with different colored measurement noises cases is simulated to validate the proposed method. Highlights • The state estimation is considered for Markovian jump nonlinear systems with colored measurement noises. • The colored measurement noise is obeyed a nonlinear autoregressive process of order n. • A new measurement equation is reconstructed by applying the statistical linear regression to the colored noise model. • The recursive posterior probability density of the state is derived and a novel adaptive Gaussian mixture filter is proposed. • The Gaussian components with small weights are pruned adaptively by using the family of Alpha (Beta) divergence measure. [ABSTRACT FROM AUTHOR]

Ten years ago, international crisis (beginning 2008) has become a fundamental element for national economies. In addition to economic burden, the crisis has put governments in an instability financial situation also, by both domestic and overseas transactions. National Governments main purpose is to provide sustainable development by attempting to prospect and plot economic evolutions in order to reduce financial influences overall. The core question is then: Are these countries able to anticipate these evolutions? This paper aims to determine how Chaos Theories analyses for Real Gross Domestic Product (GDP) and Consumption per Capita influence the trend of these indicators, but not in the classic macroeconomic dependency direction. The purpose of this article is to provide a core framework from the Chaos Theory perspective in order to create the context for Auto Regressive and Mobile Average (ARMA Models) alternatives methodology in practice, alternatives, and forecasts. The estimations based on econometric and cybernetic models by using ARMA and Hurst Coefficient highlight the fact that Real GDP and Consumption per capita have similar evolutions, but are not directly correlated in the classic theory because here, similarity and persistency are main factors for general evolutions of these macroeconomic indicators. [ABSTRACT FROM AUTHOR]

Objective In this paper, we proposed a new feature extraction approach based on the multivariate auto regressive model (MVAR) model of the sensitive intrinsic mode function (IMF) groups in the multivariate empirical mode decomposition (MEMD) domain. Approach We computed eigen values from the coefficient matrix of the MVAR model for classifying three different non-motor cognitive task in EEG based brain computer interface (BCI) system. In the first stage, the application of MEMD to multichannel EEG data gave rise to adaptive i.e. data driven decomposition of the multivariate time series data into a large number of IMF groups. In the second stage, the sensitive IMF groups were selected according to their task correlation factor. MVAR model of order six was developed from the five sensitive IMF groups and finally the eigen values of the correlation matrix derived from the coefficient matrix was employed for forming the feature vectors. At the last stage, the extracted feature vectors were fed to a Least Squares Support Vector Machine (LS-SVM) classifier for automatic classification of mental task EEG signals. We tested our approach on the mental task EEG data sets of three subjects. Main result We achieved highest value of average classification accuracy of 94.43% for binary classification of the first pair of mental task i.e baseline and mental arithmetic task using polynomial kernel and 91.65% for the second pair i.e mental arithmetic and mental letter composing task using radial basis function (RBF) with ten fold cross validation. We achieved highest value of average classification accuracy of 77.77% for three class classification employing One Vs One scheme of multiclass SVM classifier. Significance The performance of the binary classifier was evaluated on various parameters such as accuracy, specificity and sensitivity. The encouraging results show the potential of the proposed approach for classifying any non linear and non-stationary signals. [ABSTRACT FROM AUTHOR]

Models based on equations with derivatives and differences of fractional order are increasingly used for simulation of different processes. The paper presents a method of structural parametric identification of Gegenbauer autoregressive with noise on output signal. The search of the parameter of Gegenbauer difference is carried out with the help of a genetic algorithm. The propose algorithm was realized in Matlab. The simulation results show the high efficiency of the propose algorithm. [ABSTRACT FROM AUTHOR]

This paper presents an individualized model predictive control (MPC) algorithm for overnight blood glucose stabilization in people with type 1 diabetes (T1D). The MPC formulation uses an asymmetric objective function that penalizes low glucose levels more heavily. We compute the model parameters in the MPC in a systematic way based on a priori available patient information. The model used by the MPC algorithm for filtering and prediction is an autoregressive integrated moving average with exogenous input (ARIMAX) model implemented as a linear state space model in innovation form. The control algorithm uses frequent glucose measurements from a continuous glucose monitor (CGM) and its decisions are implemented by a continuous subcutaneous insulin infusion (CSII) pump. We provide guidelines for tuning the control algorithm and computing the Kalman gain in the linear state space model in innovation form. We test the controller on a cohort of 100 randomly generated virtual patients with a representative inter-subject variability. We use the same control algorithm for a feasibility overnight study using 5 real patients. In this study, we compare the performance of this control algorithm with the patient's usual pump setting. We discuss the results of the numerical simulations and the in vivo clinical study from a control engineering perspective. The results demonstrate that the proposed control strategy increases the time spent in euglycemia. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of online hop timing detection and frequency estimation of multiple frequencyhopping (FH) signals with antenna arrays. The problem is deemed as a dynamic one, as no information about the hop timing, pattern, or rate is known in advance, and the hop rate may change during the observation time. The technique of particle filtering is introduced to solve this dynamic problem, and real-time frequency and direction of arrival estimates of the FH signals can be obtained directly, while the hop timing is detected online according to the temporal autoregressive moving average process. The problem of network sorting is also addressed in this paper. Numerical examples are carried out to show the performance of the proposed method. [ABSTRACT FROM AUTHOR]

Abstract: Hedge funds have increased dramatically last decade to be the main force of the global financial market. In this paper, we mainly study the interactions between hedge funds and the world''s major stock markets to infer the law of international capital flow. We select ten international stock markets consisting of emerging and developed markets as research objects. Firstly, it shows that the stock markets of Australia, United Kingdom and the United States are highly connected with hedge fund whereas situations are different for the stock markets of China and Brazil through cross correlation analysis from 2000 to 2008. Secondly, we examine if there exists dynamic long-term relationship between the two by the method of unit root tests and co-integration tests. Finally, based on Granger causality test in a vector autoregressive process, this paper empirically analyzes how the interactions changed before and after the subprime crisis. The results show that hedge funds were influenced deeply by the subprime crisis and the industry has changed their trading positions in the stock markets a lot. [Copyright &y& Elsevier]

Shear structures are used to model tall buildings which play important roles in supporting people's living and working in large cities. These buildings will be gradually deteriorated with the material aging and environmental corrosions. In this paper, an innovative substructure approach is proposed to locate and quantify damages in any storeys of a shear structure with only using three accelerometers. A multiple partition technology is used to extract physical parameters such as stiffness and damping of single inter-storey connection from an independent substructure only containing one storey cut out from the whole structure. By doing this multiple division, damages happened in multi storeys can be identified separately which is a big progress compared to previous researches that mainly consider one damaged storey in whole structure. As in the previous researches, an autoregressive-moving average with exogenous inputs (ARMAX) model is introduced to extract the modal information of each lower substructure and to encapsulate the complex input information on the boundaries of substructure, which can effectively reduce the number of sensors and the calculation process. A damage indicator combining both the information of ARMAX model parameters and model residuals is utilized to enhance the damage identification robustness, where the Kolmogorov-Smirnov (K-S) statistic test is used to evaluate the difference between ARMAX model residuals of tested substructure and of referenced substructure. Simulation of a 10-storey shear structure and an experiment of a 5-storey shear structure have been conducted to verify the effectiveness of the proposed approach, which demonstrate that the innovative approach can accurately identify the structural damages in multi storeys even with large environmental noise disturbance. [ABSTRACT FROM AUTHOR]

According to literature and, after all, to the economic rationale, the availability and quality of the human resources is a critical factor for vivid, flourishing and pervasive research, development and innovation (RDI) activities. Indeed, probably more than in many other social and economic areas, in the RDI sector, the human capital is hardly a fungible input. As research and innovation stand out as a prerequisite for economic growth, for higher competitiveness, as well as for the society's general welfare, the human resources in RDI become a very important asset for economic, social and environmental sustainability and resilience. This paper looks into the availability of the human capital for research, from the perspective of the attractiveness of the Romanian RDI system for the potential human resources. Drawing on literature, on qualitative analysis, and applying statistical tools such as autoregressive models built on panel data available for the EU countries, the authors identify and assess the relevance of the main factors considered most important for the capacity of a RDI system to attract and keep the most valuable human resources. They also draw a few strategic action lines that may boost the interest of the qualified individuals, for research and academic careers. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for improving performances of magnetic levitation system. Controlled parameter is the amplitude which levitated object achieves during movement from one levitation position to another. Two position levitation with improved amplitude performances is obtained by implementing orthogonal neural network in standard levitation control logic. Proposed network is a nonlinear autoregressive neural network with newly developed activation function based on orthogonal polynomials. Performed experiments on a system with default control logic showed that it could not provide stable two position levitation when specified amplitude of the levitation object is greater than 10-4 m. Artificial network was trained using real experimental data and it was based on standard tangent and sigmoid activation functions. Default activation functions were then substituted with a newly developed orthogonal polynomial functions. The amplitude 10-3 m was achieved with stable two position levitation after parameter optimization. It is proven that simple control logic with nonlinear autoregressive neural network and proper activation function can provide improved amplitude performances. [ABSTRACT FROM AUTHOR]

This paper introduces an autoregressive hidden Markov model (HMM) and demonstrates its application to the speech signal. In this variant of the HMM the observed signal is assumed to be Gaussian autoregressive and the probability density function is derived based on an approximation of the linear prediction error. A Baum-Welch style set of re-estimation formulas are then derived and used to infer the model parameters for a given data set, which correspond to linguistic structure in the context of speech data. The new set of re-estimation formulas are then applied to speech data and experimental results demonstrate inference of broad phonetic categories without prior knowledge of linguistic information. The experimental results and stability of this model are then briefly contrasted with historic experiments wherein phonetic information has been inferred directly from the speech signal using a similar autoregressive model. [ABSTRACT FROM AUTHOR]

The aim of this paper is to study a possible connection between the characteristics of the labour market, namely, employment protection and flexibility, and the dynamics of the housing market. More specifically, this contribution analyses whether the poor quality of employment in a given economy could cause the collapse of the housing market and impairs its recovery. In the second stage of our analysis, we provide estimates for Spain over the period 1984-2014. In doing so, the autoregressive distributed lag (ARDL) bounds test for cointegration is employed. [ABSTRACT FROM AUTHOR]

PieceWise AutoRegressive eXogenous (PWARX) models represent one of the broad classes of the hybrid dynamical systems (HDS). Among many classes of HDS, PWARX model used as an attractive modeling structure due to its equivalence to other classes. This paper presents a novel fuzzy distance weight matrix based parameter identification method for PWARX model. In the first phase of the proposed method estimation for the number of affine submodels present in the HDS is proposed using fuzzy clustering validation based algorithm. For the given set of input–output data points generated by predefined PWARX model fuzzy c-means (FCM) clustering procedure is used to classify the data set according to its affine submodels. The fuzzy distance weight matrix based weighted least squares (WLS) algorithm is proposed to identify the parameters for each PWARX submodel, which minimizes the effect of noise and classification error. In the final phase, fuzzy validity function based model selection method is applied to validate the identified PWARX model. The effectiveness of the proposed method is demonstrated using three benchmark examples. Simulation experiments show validation of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, a detailed study on the possibility and significance of performing a parametric estimation of sample entropy (SampEn) is proposed. SampEn is a non-linear metric, meant to quantify regularity of a time series. It is widely employed on biomedical signals, especially on heart rate variability. Results relevant to approximate entropy, a related index, are also reported. An analytical expression for SampEn of an autoregressive (AR) model is derived first. Then we study the feasibility of a parametric estimation of SampEn through AR models, both on synthetic and real series. RR series of different lengths are fitted to an AR model and then expected values of SampEn (SampEn μ ) are estimated. Values of SampEn, computed from real beat-to-beat interval time series (obtained from 72 normal subjects and 29 congestive heart failure patients), with m = 1 and r = 0.2, are within the standard range of SampEn μ more than 83% (for series length N = 75) and 28% (for N = 1500) of the cases. Surrogate data have been employed to verify if departures from Gaussianity are to account for the mismatch. The work supports the finding that when numerical and parametric estimates of SampEn agree, SampEn is mainly influenced by linear properties of the series. A disagreement, on the contrary, might point those cases where SampEn is truly offering new information, not readily available with traditional temporal and spectral parameters. [ABSTRACT FROM AUTHOR]

• Brain computer interface (BCI) can help human to control their environment by robots and machine. • Motor imaginary detection as a part of BCI, is used for extracting imaginary orders from brain for order classification. • Electroencephalography is one of the modalities for motor imaginary classification. • Paper proposed a method based on dual Kalman filter and visibility graphs form source activity of the brain. • Proposed method gives better accuracy and classification rate in comparison to other methods. Nowadays Brain Computer Interface (BCI) is one of the most important fields in neuroscience in which machine works are controlled with the human brain. Motor imagery recognition is one part of BCI study in which several papers are published in this field. In this paper, four individual motor imagery (left and right hand, foot, and tongue) are classified by using visibility graph features of brain source dynamic which are extracted from EEG signals. In the proposed method, firstly EEG signals are recorded and preprocessed. Source localization algorithm (sLORETA in this paper) is applied to EEG signals and active sources during time are extracted. Then the multivariate autoregressive model (MVAR) is fitted to active sources and dual Kalman filter is applied to MVAR model for source dynamics and their relations estimation. Then extracted source dynamic signals are mapped to 5 different visibility graphs and some features are extracted from each graph. Extracted features number are reduced by analysis of variance (ANOVA) method and support vector machine (SVM) classifiers are applied for motor imagery classification. In this research, the competition IV of BCI database was used and the results indicate that the proposed algorithm has a superior performance to the other methods. By classifying motor imaginary tasks with source dynamic signals better results with more accuracy and good performance measures are achieved in comparison with EEG signals classification by extracted from visibility graph gives better performance between graphs. [ABSTRACT FROM AUTHOR]

This paper presents an on-line myoelectric control system which can classify eight prehensile hand gestures with only two electrodes. The overlapping windowing scheme is adopted in the system leading a continuous decisions flow. We choose mean absolute value (MAV), variance (VAR), the fourth-order autoregressive (AR) coefficient and Sample entropy (SampEn) as the feature set and utilize the linear discriminant analysis (LDA) to reduce the dimension and obtain the projected feature sets. The current projected feature set and the previous one are "pre-smoothed" before the classification, and then a decision is generated by LDA classifier. To get the final decision from the decisions flow, the current decision and m previous decisions are "post-smoothed". The method mentioned above can obtain a 99.04% off-line accuracy rate and a 97.35% on-line accuracy rate for individual gesture. By choosing a proper value of m, this method can also get a 99.79% accuracy rate for on-line recognition of complex sequences of hand gestures without interruption. In addition, a virtual hand has been developed to display the on-line recognition result visually, and a proper control strategy is proposed to realize the continuous switch of hand gestures. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes a new and effective framework for color image retrieval based on Full Range Autoregressive Model (FRAR). Bayesian approach (BA) is used to estimate the parameters of the FRAR model. The color autocorrelogram, a new version of edge histogram descriptor (EHD) and micro-texture (MT) features are extracted using a common framework based on the FRAR model with BA. The extracted features are combined to form a feature vector, which is normalized and stored in image feature vector database. The feature vector database is categorized according to the nature of the images using the radial basis function neural network (RBFNN) and k-means clustering algorithm. The proposed system adopted Manhattan distance measure of order one to measure the similarity between the query and target images in the categorized and indexed feature vector database. The query refinement approach of short-term learning based relevance feedback mechanism is adopted to reduce the semantic gap. The experimental results, based on precision and recall method are reported. It demonstrates the performance of the improved EHD, effectiveness and efficiency achieved by the proposed framework. [Copyright &y& Elsevier]

• Quantifying alertness intuitively. • Monitoring EEG changes during continuous tasks. • Better classification performance (with better accuracy and precision) • Applying decision fusion on the BP neural network. Alertness detection has been used to monitor operators in key positions to avoid accidents. The widely used methods (such as machine vision) are susceptible to external factors and cannot quantify alertness. In contrast, EEG has the advantages of intuitively reflecting brain activities and not being easily affected. Based on EEG signals, an alertness detection method using decision fused BP (back propagation, BP) neural network was presented. In this paper, we analyzed the EEG changes of 20 participants while performing tasks. During PVT (Psychomotor Vigilance Task, PVT), EEG signals were collected through the acquisition platform we designed, and KSS (Karolinska Sleepiness Scale, KSS) was completed. AR (autoregressive model, AR) and PSD (power spectral density, PSD) were applied to extract frequency domain features of EEG signals. The extracted features were then input into BP neural network separately. Afterwards, we used DS (Dempster–Shafer, DS) method to fuse output results of BP neural network. When comparing results of the proposed classification methods, we concluded that BP-DS method had the finest classification results (including AUC-ROC curve, accuracy, precision, and recall rate) with the accuracy increasing considerably from 0.661 to 0.911. In addition, among six EEG indices, R β , α + θ / β , α / β presented prominent changes during alertness variation. The proposed alertness classification algorithm monitors alertness continuously and avoids potential catastrophes caused by decreased alertness. It can be applied in monitoring systems to prevent mental fatigue and declined work efficiency of air traffic controllers, medical staff and other fields. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we propose the large margin autoregressive (LMAR) model for classification of time series patterns. The parameters of the generative AR models for different classes are estimated using the margin of the boundaries of AR models as the optimization criterion. Models that use a mixture of AR (MAR) models are considered for representing the data that cannot be adequately represented using a single AR model for a class. Based on a mixture model representing each class, we propose the large margin mixture of AR (LMMAR) models. The proposed methods are applied on the simulated time series data, electrocardiogram data, speech data for E-set in English alphabet and electroencephalogram time series data. Performance of the proposed methods is compared with that of support vector machine (SVM) based classifier that uses AR coefficients based features. The proposed methods give a better classification performance compared to the SVM based classifier. Being generative models, the LMAR and LMMAR models provide a generative interpretation that enables utilization of the rejection option in the high risk classification tasks. The proposed methods can also be used for detection of novel time series data. [Copyright &y& Elsevier]

Abstract: In this paper we present two methodologies to generate heart rate variability (HRV) signals characterized by controlled and real-like time-frequency (TF) structure to be used to assess different methods of non-stationary HRV analysis. The synthesized signals are stochastic processes whose TF structure is predetermined by choosing either the time-course of the instantaneous frequencies and powers or the shape of the TF model function. They consist of three steps: (a) choice of the desired TF structure of the signals by choosing a set of design parameters; (b) automatic identification of the parameters of the corresponding models via simple closed-form expressions; (c) synthesis of the desired stochastic signals. Two measures to evaluate the goodness of the simulated signals are also given. Using this framework we were able to model the wide range of non-stationarities observed in heart rate modulation during exercise stress testing and experiments of music-induced emotions. We used the proposed methodology to assess the capability of the smoothed pseudo Wigner–Ville distribution (SPWVD) to quantify HRV patterns. We observed that the SPWVD followed the temporal evolution of the spectral components even when sudden and sharp transitions occur. [Copyright &y& Elsevier]

• Generalized harmonic analysis is used for decomposing harmonics of heart rate signal. • Compared with autoregressive model or Pisarenko harmonic decomposition. • Useful tool for calculating the power of a component. • The limitations of generalized harmonic analysis are still discussed. In this paper, a method for decomposing harmonics in the spectrum of heart rate variability (HRV) using generalized harmonic analysis (GHA) is introduced. First, a simulated RR interval signal generated by an integral pulse frequency modulation model was decomposed spectra by GHA, Pisarenko harmonic decomposition (PHD), and autoregressive (AR) spectral decomposition model. The spectral profiles were obtained by the GHA, PHD, and AR methods for various numbers of extracted sinusoids and model orders from 1 to 48. The spectral profiles of GHA were the most stable. Of the power values of the sinusoids extracted by each method, it was clear that the power values estimated by GHA were approximately equal to the mean square value and closer than that obtained using the PHD or a fast Fourier transform (FFT). Second, a comparison of the power of the low-frequency (LF) and high-frequency (HF) band components reveals that the values obtained by GHA are similar to those obtained by FFT for the analysis using a real ECG signal. Third, Bland-Altman analysis reveals that LF and HF band power value calculated by the GHA are compatible with ones by the Lomb-Scargle periodogram using MIT-BIH normal sinus rhythm database from short term recordings of 30 min. These results suggest that GHA is a useful tool for calculating the power of a component in the spectral analysis of HRV. The limitations of spectral decomposition by GHA are still discussed. [ABSTRACT FROM AUTHOR]

• Graph signal processing is employed for EEG-based motor imagery classification. • Multichannel EEG graph models with direction and weight information are built. • A new graph Fourier transform is defined with a symmetric graph Laplacian. • Results on two BCI datasets exceed (or match) existing state-of-the-art results. Motor imagery patterns are extensively exploited in brain-computer interface systems in order to control outer devices without using peripheral nerves or muscles. Classification of these patterns can be based on the associated electroencephalogram (EEG) signals. Recent approaches addressed this classification problem through techniques exploiting mainly information from one or two EEG channels. However, these approaches overlook correlations between multiple EEG channels. In this paper, we create motor-imagery classification systems based on graph-theoretic models of multichannel EEG signals. In particular, multivariate autoregressive models are used to establish the relations between the EEG channels and construct directed graph signals. Also, we constructed undirected graph signal models with Gaussian-weighted distances between graph nodes. Then, a novel variant of the graph Fourier transform is applied to the directed and undirected graph models with and without edge weights. Distinctive features were thus extracted from the transform coefficients. Additional features were computed using common spatial patterns, polynomial representations and principal components of EEG signals. Significant performance improvements were achieved using extreme learning machine (ELM) classifiers. For Dataset Ia of the BCI Competition 2003, our approach led to a classification accuracy of 96.58% with fully-connected weighted directed graph features computed on the delta-band EEG signals. For the six subjects of the Dataset 1 of the BCI Competition IV, our approach compared well with other state-of-the-art methods in the alpha and beta EEG bands. [ABSTRACT FROM AUTHOR]