Article
Engineering, Electrical & Electronic
Ali A. Radhi, Hikmat N. Abdullah, Hanan A. R. Akkar
Summary: This paper presents an intelligent centralized-cooperative spectrum sensing technique (JBKMS) based on Jarque-Bera (JB) statistics test and K-Means (KMS) clustering algorithm to develop new denoised features. The proposed method improves sensing performance by utilizing high-order moments and minimizing interference. Simulation results demonstrate superior detection performance at low signal-to-noise ratio.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Multidisciplinary Sciences
Tanweer Ul Islam
Summary: Comparison of normality tests based on absolute or average powers often leads to ambiguous results due to the critical dependence of these statistics on alternative distributions. This study proposes a new computational benchmark method to evaluate normality tests with reduced cost and accurate results.
Article
Multidisciplinary Sciences
Zhen Meng, Zhenzhen Jiang
Summary: Testing whether data are from a normal distribution is a traditional problem in data analysis and is crucial for many statistical methods. There are various tests available, such as Anderson-Darling test, Shapiro-Wilk test, and Jarque-Bera test, but they have limitations in dealing with complex and diverse data distributions. Therefore, we propose a robust and valid Cauchy Combination Omnibus Test (CCOT) that shows good properties in analyzing data from different distributions.
Article
Statistics & Probability
Wanfang Chen, Marc G. Genton
Summary: The study focuses on the application of multivariate normality tests in spatial data. Recent advances in multivariate normality tests for i.i.d. data are reviewed, and a new method that accounts for spatial dependence is proposed for spatial data.
INTERNATIONAL STATISTICAL REVIEW
(2023)
Article
Business, Finance
Ariston Karagiorgis, Konstantinos Drakos
Summary: This study examines the properties of the S-K plane for hedge funds and finds significant variations across different investment strategies.
JOURNAL OF INTERNATIONAL FINANCIAL MARKETS INSTITUTIONS & MONEY
(2022)
Article
Business, Finance
Ariston Karagiorgis, Antonis Ballis, Konstantinos Drakos
Summary: Cryptocurrency returns show significant divergence from normality, with the relationship between Skewness and Kurtosis following a parabolic form, although this connection is sparsely documented. In this study, we illustrate the characteristics of the S-K plane for cryptocurrencies using diagrams. Additionally, utilizing the data's panel structure, we estimate a quadratic model for the S-K plane. Furthermore, we investigate the impact of cryptocurrency type, infrastructure, and the time period analyzed on the architecture of the plane. Our findings indicate that the squared Skewness of tokens significantly reduces the slope of Kurtosis, particularly during the earlier stages of the market.
INTERNATIONAL JOURNAL OF FINANCE & ECONOMICS
(2023)
Article
Business, Finance
Ariston Karagiorgis, Konstantinos Drakos
Summary: Using hedge fund data, we estimate transition matrices for analyzing the mobility properties of skewness, kurtosis, and joint transitions. These matrices provide probabilistic structures of the higher moments individually and jointly. We also apply various indices to analyze the dynamics and validate our findings using Probit models. Our results show a high level of mobility towards non-normality in both investment strategies and the sector as a whole, with only a limited number of strategies maintaining normality in the joint framework.
RESEARCH IN INTERNATIONAL BUSINESS AND FINANCE
(2023)
Article
Mathematics
Jimmy Reyes, Mario A. Rojas, Pedro L. Cortes, Jaime Arrue
Summary: This paper introduces a new family of distributions based on the Laplace distribution. The new family is defined by its stochastic representation as the sum of two independent random variables, one with a Laplace distribution and the other with an exponential distribution. The statistical performance of the estimators obtained by the moments and maximum likelihood methods were empirically evaluated using Monte Carlo simulation study. The results show that the exponentially modified Laplace model can be used as an alternative distribution to model skewed data with high kurtosis.
Article
Computer Science, Interdisciplinary Applications
Hyowon An, Kai Zhang, Hannu Oja, J. S. Marron
Summary: Identification of important variables in big data is a crucial challenge. To tackle this, methods for discovering variables with non-standard univariate marginal distributions are proposed. Traditional moments-based summary statistics can be sensitive to outliers, thus L-moments are considered for robustness. However, the limitation of L-moments is addressed by proposing Gaussian Centered L-moments.
COMPUTATIONAL STATISTICS & DATA ANALYSIS
(2023)
Article
Economics
Carol Alexander, Emese Lazar, Silvia Stanescu
Summary: This study derives the conditional moments of the GJR-GARCH model, analyzes the limits of moments as the time horizon increases, and validates the results through simulation and empirical studies.
INTERNATIONAL JOURNAL OF FORECASTING
(2021)
Article
Business, Finance
Walid M. A. Ahmed, Mustafa Al Mafrachi
Summary: This study finds that cryptocurrency returns are sensitive to higher-order realized moments, with skewness and hyper-skewness showing statistically significant predictive capabilities for future returns. The impact of these moments on cryptocurrency returns varies depending on the type and state of the cryptomarket.
INTERNATIONAL REVIEW OF ECONOMICS & FINANCE
(2021)
Article
Business, Finance
Austin Shelton, Hayden Kane, Charles Favreau
Summary: Researchers developed a novel method to estimate a stock's risk-neutral return moments, but errors could be introduced in practice. They recommend using different methodologies to estimate implied moments for more accurate results.
QUANTITATIVE FINANCE
(2021)
Article
Business, Finance
Pierpaolo Uberti
Summary: This paper proposes a generalization of the Markowitz model by incorporating skewness and kurtosis into the classical mean-variance allocation framework. The key advantage is that it provides a closed-form solution to the optimization problem. The four moments optimal portfolio is decomposed into three portfolios: the mean-variance optimal portfolio and two self-financing portfolios, which account for skewness and kurtosis, respectively. The paper discusses the theoretical properties of the optimal solution and provides an economic interpretation. Finally, an empirical exercise with real financial data illustrates the contribution of the two portfolios considering skewness and kurtosis when financial returns deviate from the Normal distribution.
QUANTITATIVE FINANCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Dequan Zhang, Junkai Jia, Zhonghao Han, Heng Ouyang, Jie Liu, Xu Han
Summary: In this study, a derivative lambda probability density function (lambda-PDF) is proposed for quantifying the uncertainties in structural uncertainty problems with nonconventional distributions. An efficient uncertainty propagation approach based on the improved derivative lambda-PDF and dimension reduction method (DRM) is developed. The proposed method shows superiority over reference methods in terms of effectiveness and accuracy.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2023)
Article
Engineering, Electrical & Electronic
Ali A. Radhi, Hanan A. R. Akkar, Hikmat N. Abdullah
Summary: This paper proposes a novel blind cooperative spectrum sensing scheme based on new denoised mixed-features and the K-Medoids algorithm to improve the poor performance of a traditional cooperative spectrum sensing system in low signal-to-noise ratio conditions. By extracting denoised mixed-features and using highest-order moments, the computational complexity is reduced, and the effect of noise uncertainty is mitigated, resulting in improved sensing performance.
PHYSICAL COMMUNICATION
(2022)