Multivariate Hyper-Rotated GARCH-BEKK

For large multivariate models of generalized autoregressive conditional heteroskedasticity (GARCH), it is important to reduce the number of parameters to cope with the ‘curse of dimensionality’. Recently, Laurent, Rombouts and Violante (2014 “Multivariate Rotated ARCH Models” Journal of Econometrics 179: 16–30) developed the rotated multivariate GARCH model, which focuses on the parameters for standardized variables. This paper extends the rotated multivariate GARCH model by considering a hyper-rotation, which uses a more flexible structure for the rotation matrix. The paper shows an alternative representation based on a random coefficient vector autoregressive and moving-average (VARMA) process, and provides the regularity conditions for the consistency and asymptotic normality of the quasi-maximum likelihood (QML) estimator for VARMA with hyper-rotated multivariate GARCH. The paper investigates the finite sample properties of the QML estimator for the new model. Empirical results for four exchange rate returns show the new specifications works satisfactory for reducing the number of parameters.

Fully Automatic Cell Segmentation with Fourier Descriptors

Cell segmentation is a fundamental problem in biology for which convolutional neural networks yield the best results nowadays. In this paper, we present HarmonicNet, a network, which is a modification of the popular StarDist and SplineDist architectures. While StarDist and SplineDist describe an object by the lengths of equiangular rays and control points respectively, our network utilizes Fourier descriptors, predicting a coefficient vector for every pixel on the image, which implicitly define the resulting segmentation. We evaluate our model on three different datasets, and show that Fourier descriptors can achieve a high level of accuracy with a small number of coefficients. HarmonicNet is also capable of accurately segmenting objects that are not star-shaped, a case where StarDist performs suboptimally according to our experiments.

Solving the dilemma in supplier selection by the group of weight vector with kernel

PurposeThe purpose of this paper is to solve the dilemma in the process of major selection decision-making.Design/methodology/approachFirstly, the group of weight vector with kernel has been defined. Then, the weighted comprehensive clustering coefficient vector was calculated based on the group of weight vector with kernel. Under the action of weighted comprehensive clustering coefficient vector, the information including in other components around component k and supporting object i to be classified into the k-th category has been gathered to component k. At last, a novel two-stage decision model based on the group of weight vector with kernel and the weighted comprehensive clustering coefficient vector is put forward to solve the dilemma in grey clustering evaluation. Then the overall evaluation conclusion can be consistent with the clustering result according to the rule of maximum value.FindingsA new way to solve the dilemma in the process of major selection decision-making has been found. People can obtain a consistent result with two-stage decision model at the case of dilemma. That is, the conclusion of the overall evaluation is consistent with the clustering result according to the rule of maximum value.Practical implicationsSeveral functional groups of weight vector with kernel have been put forward. The proposed model can solve the clustering dilemma effectively and produce consistent results. A practical application of decision problem to solve the dilemma in supplier evaluation and selection of a key component of large commercial aircraft C919 have been completed by the novel two-stage decision model.Originality/valueThe two-stage decision model, the group of weight vector with kernel and the weighted comprehensive clustering coefficient vector were presented in this paper firstly. People can solve the dilemma in grey clustering evaluation effectively by the novel two-stage decision model based on the group of weight vector with kernel and the weighted comprehensive clustering coefficient vector.

Prior Assumptions for Leak Localisation in Water Distribution Networks with Uncertainties

Hydraulic model-based leak (burst) localisation in water distribution networks is a challenging problem due to a limited number of hydraulic measurements, a wide range of leak properties, and model and data uncertainties. In this study, prior assumptions are investigated to improve the leak localisation in the presence of uncertainties. For example, $$\ell _2$$ ℓ 2 -regularisation relies on the assumption that the Euclidean norm of the leak coefficient vector should be minimised. This approach is compared with a method based on the sensitivity matrix, which assumes the existence of only a single leak. The results show that while the sensitivity matrix method often yields a better leak location estimate in single leak scenarios, the $$\ell _2$$ ℓ 2 -regularisation successfully identifies a search area for pinpointing the accurate leak location. Furthermore, it is shown that the additional error introduced by a quadratic approximation of the Hazen-Williams formula for the solution of the localisation problem is negligible given the uncertainties in Hazen-Williams resistance coefficients in operational water network models.

Fourth kind Chebyshev Wavelet Method for the solution of multi-term variable order fractional differential equations

PurposeMulti-term variable-order fractional differential equations (VO-FDEs) are powerful tools in accurate modeling of transient-regime real-life problems such as diffusion phenomena and nonlinear viscoelasticity. In this paper the Chebyshev polynomials of the fourth kind is employed to obtain a numerical solution for those multi-term VO-FDEs.Design/methodology/approachTo this end, operational matrices for the approximation of the VO-FDEs are obtained using the Fourth kind Chebyshev Wavelets (FKCW). Thus, the VO-FDE is condensed into an algebraic equation system. The solution of the system of those equations yields a coefficient vector, the coefficient vector in turn yields the approximate solution.FindingsSeveral examples that we present at the end of the paper emphasize the efficacy and preciseness of the proposed method.Originality/valueThe value of the paper stems from the exploitation of FKCWs for the numerical solution of multi-term VO-FDEs. The method produces accurate results even for relatively small collocation points. What is more, FKCW method provides a compact mapping between multi-term VO-FDEs and a system of algebraic equations given in vector-matrix form.

SAR Target Recognition Using Improved Sparse Representation with Local Reconstruction

In order to handle the problem of synthetic aperture radar (SAR) target recognition, an improved sparse representation-based classification (SRC) is proposed. According to the sparse coefficient vector resulting from the global dictionary, the largest coefficient in each class is taken as the reference. Then, the surrounding neighborhoods of the sample with the largest coefficient are selected to construct the optimal local dictionary in each training class. Afterwards, the samples in the local dictionary are used to reconstruct the test sample to be identified. Finally, the decision is made according to the comparison of the reconstruction errors from different classes. In the experiments, the proposed method is verified based on the moving and stationary target acquisition and recognition (MSTAR) dataset. The results show that the proposed method has performance advantages over existing methods, which demonstrates its effectiveness and robustness.

The Role of Prior Assumptions for Leak Localisation in Water Networks with Uncertainties

Hydraulic model-based leak (burst) localisation in water networks is a challenging problem due to uncertainties, the limited number of hydraulic measurements, and the wide range of leak properties. In this study, we investigate the use of prior assumptions to improve the leak localisation in the presence of model uncertainties. For example, 𝓁2-regularisation relies on the assumption that the Euclidean norm of the leak coefficient vector should be minimised. This approach is compared with a method based on the sensitivity matrix, which assumes the existence of only a single leak. We show that while applying the sensitivity matrix often yields a better estimate of the leak location in single leak scenarios, the 𝓁2-regularisation successfully identifies a leak search area for pinpointing the accurate leak location. Furthermore, we demonstrate that the additional error introduced by a quadratic approximation of the Hazen-Williams formula for the solution of the localisation problem is negligible given the uncertainties in Hazen-Williams resistance coefficients in operational water network models.

A New Stochastic Model Updating Method Based on Improved Cross-Model Cross-Mode Technique

This paper proposes a new stochastic model updating method to update structural models based on the improved cross-model cross-mode (ICMCM) technique. This new method combines the stochastic hybrid perturbation-Galerkin method with the ICMCM method to solve the model updating problems with limited measurement data and uncertain measurement errors. First, using the ICMCM technique, a new stochastic model updating equation with an updated coefficient vector is established by considering the uncertain measured modal data. Then, the stochastic model updating equation is solved by the stochastic hybrid perturbation-Galerkin method so as to obtain the random updated coefficient vector. Following that, the statistical characteristics of the updated coefficients can be determined. Numerical results of a continuous beam show that the proposed method can effectively cope with relatively large uncertainty in measured data, and the computational efficiency of this new method is several orders of magnitude higher than that of the Monte Carlo simulation method. When considering the rank deficiency, the proposed stochastic ICMCM method can achieve more accurate updating results compared with the cross-model cross-mode (CMCM) method. An experimental example shows that the new method can effectively update the structural stiffness and mass, and the statistics of the frequencies of the updated model are consistent with the measured results, which ensures that the updated coefficients are of practical significance.

Oracle inequalities for weighted group lasso in high-dimensional misspecified Cox models

We study the nonasymptotic properties of a general norm penalized estimator, which include Lasso, weighted Lasso, and group Lasso as special cases, for sparse high-dimensional misspecified Cox models with time-dependent covariates. Under suitable conditions on the true regression coefficients and random covariates, we provide oracle inequalities for prediction and estimation error based on the group sparsity of the true coefficient vector. The nonasymptotic oracle inequalities show that the penalized estimator has good sparse approximation of the true model and enables to select a few meaningful structure variables among the set of features.

Comparison of Two Estimators of the Regression Coefficient Vector Under Pitman’s Closeness Criterion

Schaffrin and Toutenburg [1] proposed a weighted mixed estimation based on the sample information and the stochastic prior information, and they also show that the weighted mixed estimator is superior to the ordinary least squares estimator under the mean squared error criterion. However, there has no paper to discuss the performance of the two estimators under the Pitman’s closeness criterion. This paper presents the comparison of the weighted mixed estimator and the ordinary least squares estimator using the Pitman’s closeness criterion. A simulation study is performed to illustrate the performance of the weighted mixed estimator and the ordinary least squares estimator under the Pitman’s closeness criterion.