Article
Business, Finance
Takuji Matsumoto, Derek Bunn, Yuji Yamada
Summary: Short-term risk management is crucial in power trading with the introduction of more weather risk by intermittent renewable generators. This paper analyzes a flexible hedging product, day-ahead cap futures, and parametrically predicts the probability distribution of day-ahead prices using the multifactor Generalized Additive Model. The higher-order moment model is shown to be superior in terms of fairness, underwriting risk, and variance reduction compared to lower-order models like the normal distribution.
QUANTITATIVE FINANCE
(2022)
Article
Multidisciplinary Sciences
Tonu Kollo, Meelis Kaarik, Anne Selart
Summary: Symmetric elliptical distributions have been widely used in data modeling and robustness studies. After transforming into skew elliptical distributions, the area of applications has been significantly expanded. The study focuses on the skew t-distribution, with special attention given to the case when mu = 0 for constructing skew t-copula.
Article
Statistics & Probability
Bader Alruwaili
Summary: This paper studies the modality of skew t-distribution and the mixture of skew t-distributions graphically and analytically. A new formula is introduced to find the modes for these distributions. The effects of skewness parameters and degrees of freedom on the number and locations of modes are explored, as well as the merging strategy for components in the mixture model.
STATISTICAL PAPERS
(2023)
Article
Mathematics
Raul Alejandro Moran-Vasquez, Edwin Zarrazola, Daya K. Nagar
Summary: The multivariate skew-t distribution is important in statistics as it combines skewness with heavy tails, which are common in real-world data. This article studies various distributional properties of the truncated multivariate skew-t distribution and provides a method for generating random samples from this distribution.
Article
Mathematics, Applied
Weisan Wu
Summary: Image segmentation technology is widely used in various business and social fields, with more scholars studying theories in this field and developing effective models and methods. However, as demand for higher quality images increases, the need for deeper understanding and analysis of image segmentation theory grows. In this study, Renyi entropy and Shannon entropy of finite multivariate skew t mixture distribution were investigated, providing better properties and wider application range than traditional skew t distribution. The use of Holder inequality and polynomial theorem helped establish upper and lower bounds for the two types of entropy in the distribution.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2021)
Article
Automation & Control Systems
Weidong Wang, Qin Yu, Maria Fasli
Summary: Gaussian process regression is commonly used to approximate the maximum distribution of random variables in Bayesian optimization, but the introduction of acquisition function can increase computational cost and complexity. A Monte Carlo-like method has been proposed as an alternative to approximate the maximum distribution, and constructing the maximum distribution within the framework of Student-t process shows competitive performance compared to the traditional GP-based method.
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Wendyam Eric Lionel Ilboudo, Taisuke Kobayashi, Kenji Sugimoto
Summary: This paper proposes a robust stochastic gradient optimization method based on the student-t distribution, in which the robustness is directly built in the algorithm. By integrating the method into several stochastic gradient algorithms, including the widely used optimizer Adam, the resultant algorithm outperforms Adam and their original versions in terms of robustness against noise across various tasks.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Statistics & Probability
Tsung- Lin, I-An Chen, Wan-Lun Wang
Summary: This paper introduces a more flexible robust factor analysis model based on the CFUST distribution, which can account for more complex features of skewness toward multiple directions. An efficient algorithm is developed for parameter estimation and an information-based approach is used to assess the variability of parameter estimates. The effectiveness and applicability of the proposed techniques are demonstrated through the analysis of simulated and real datasets.
STATISTICAL PAPERS
(2023)
Article
Statistics & Probability
Pradnya P. Khandeparkar, V. U. Dixit
Summary: The paper discusses the Skew log Laplace distribution and the generalised skew log Laplace distribution for analysing reported incomes that are mixtures of true and underreported incomes. It presents the maximum likelihood estimators, confidence intervals, and a likelihood ratio test for choosing between the distributions. An application of the distributions to real life income values is also provided.
JOURNAL OF THE INDIAN SOCIETY FOR PROBABILITY AND STATISTICS
(2021)
Article
Mathematics
Jaime Arrue, Reinaldo B. Arellano-Valle, Enrique Calderin-Ojeda, Osvaldo Venegas, Hector W. Gomez
Summary: In this paper, likelihood-based inference and bias correction using Firth's approach are developed for the modified skew-t-normal (MStN) distribution. Compared to the modified skew-normal (MSN) distribution, the MStN model allows for modeling heavily skewed data and thick tails. The shape parameter and the degrees of freedom control the tails. The paper provides the density and important properties of this new distribution and presents the Fisher's information matrix and observed matrix associated with the log-likelihood. Additionally, the non-singularity of the Fisher's information matrix for the MStN model is demonstrated when the shape parameter is zero. Firth's method for bias reduction is applied to the MStN model to address the inferential problem in the shape parameter, both in scalar and location-scale cases.
Article
Engineering, Multidisciplinary
Chen Xu, Yawen Mao, Hongtian Chen, Hongfeng Tao, Fei Liu
Summary: This paper introduces a new nonlinear filtering algorithm using skew t distribution to deal with asymmetry of measurement noise. The joint estimation of system states and statistics of skew t noise distribution is conducted using variational Bayesian inference. Simulation results show that the proposed nonlinear filter performs well even when facing unknown statistics of measurement noise, outperforming existing state-of-the-art nonlinear filters.
CMES-COMPUTER MODELING IN ENGINEERING & SCIENCES
(2022)
Article
Mathematics, Applied
Se Yoon Lee
Summary: Estimating the number of degrees of freedom is challenging due to the leptokurtic nature of the Student t-distribution, especially in studies with small sample sizes. This article examines popular priors used in the literature and proposes a log-normal prior as a viable option, demonstrating its superiority through simulation studies and financial applications.
Article
Geochemistry & Geophysics
Li Liang, Qiqi He, Hualong Cao, Yang Yang, Xuefeng Chen, Gang Lin, Mina Han
Summary: This work presents a novel method for remote sensing image registration based on a dual-features Student-t distribution mixture model, which utilizes guided image filter to enhance characteristics of feature points. Experimental results demonstrate the superior performance of this method compared to five state-of-the-art methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Medicine, General & Internal
Bilal Ahmad, Sun Jun, Vasile Palade, Qi You, Li Mao, Mao Zhongjie
Summary: The paper introduces a new framework TED-GAN utilizing variational autoencoder and generative adversarial networks to generate realistic skin lesion images and enhance classification performance.
Correction
Statistics & Probability
Mahmood Kharrati-Kopaei
Summary: Celik and Senoglu (2019) proposed a two-way analysis of variance (ANOVA) model with skew t-distributed error terms. They obtained estimators for the model parameters using maximum likelihood (ML) and modified maximum likelihood (MML) methodologies. Test statistics based on these estimators were also proposed for hypothesis testing in the two-way ANOVA model. Through a Monte Carlo simulation study, they compared the efficiencies of ML and MML estimators and the power of the test statistics with normal theory results. Unfortunately, there was an error in presenting one of the maximum likelihood equations in the paper, which raises questions about some of the simulation results and conclusions.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2023)
Article
Gastroenterology & Hepatology
Salvatore Petta, Giada Sebastiani, Mauro Vigano, Javier Ampuero, Vincent Wai-Sun Wong, Jerome Boursier, Annalisa Berzigotti, Elisabetta Bugianesi, Anna Ludovica Fracanzani, Calogero Camma, Marco Enea, Marraud des Grottes, Vito Di Marco, Ramy Younes, Aline Keyrouz, Sergio Mazzola, Yuly Mendoza, Grazia Pennisi, Manuel Romero-Gomez, Antonio Craxi, Victor de Ledinghen
Summary: The study found that in patients with NAFLD and compensated advanced chronic liver disease, baseline liver stiffness measurements and their changes can be used to identify patients at risk for liver-related events and mortality.
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY
(2021)
Review
Gastroenterology & Hepatology
Arnaldo Amato, Emanuele Sinagra, Ciro Celsa, Marco Enea, Andrea Buda, Filippo Vieceli, Lucia Scaramella, Paul Belletrutti, Alessandro Fugazza, Calogero Camma, Franco Radaelli, Alessandro Repici, Andrea Anderloni
Summary: This meta-analysis showed that lumen-apposing metal stents (LAMS) and self-expandable metal stents (SEMS) are comparable in terms of efficacy for endoscopic ultrasound-guided choledochoduodenostomy (EUS-CDS). Clinical and technical success, post-procedure adverse events, and reintervention rates were similar between LAMS and SEMS use; however, adverse events require further investigation.
Article
Multidisciplinary Sciences
Bianca Magro, Valentina Zuccaro, Luca Novelli, Lorenzo Zileri, Ciro Celsa, Federico Raimondi, Mauro Gori, Giulia Camma, Salvatore Battaglia, Vincenzo Giuseppe Genova, Laura Paris, Matteo Tacelli, Francesco Antonio Mancarella, Marco Enea, Massimo Attanasio, Michele Senni, Fabiano Di Marco, Luca Ferdinando Lorini, Stefano Fagiuoli, Raffaele Bruno, Calogero Camma, Antonio Gasbarrini
Summary: A simple clinical prediction rule was developed to identify in-hospital mortality risk for COVID-19 patients. With seven independent risk factors, the prediction rule showed good discrimination and calibration in both the derivation and validation cohorts.
Article
Gastroenterology & Hepatology
Victor Sapena, Marco Enea, Ferran Torres, Ciro Celsa, Jose Rios, Giacomo Emanuele Maria Rizzo, Pierre Nahon, Zoe Marino, Ryosuke Tateishi, Tatsuya Minami, Angelo Sangiovanni, Xavier Forns, Hidenori Toyoda, Stefano Brillanti, Fabio Conti, Elisabetta Degasperi, Ming-Lung Yu, Pei-Chien Tsai, Kevin Jean, Mohamed El Kassas, Hend Ibrahim Shousha, Ashraf Omar, Claudio Zavaglia, Hiroko Nagata, Mina Nakagawa, Yasuhiro Asahina, Amit G. Singal, Caitlin Murphy, Mohamed Kohla, Chiara Masetti, Jean-Francois Dufour, Nicolas Merchante, Luisa Cavalletto, Liliana L. C. Chemello, Stanislas Pol, Javier Crespo, Jose Luis Calleja, Rosanna Villani, Gaetano Serviddio, Alberto Zanetto, Sarah Shalaby, Francesco Paolo Russo, Rob Bielen, Franco Trevisani, Calogero Camma, Jordi Bruix, Giuseppe Cabibbo, Maria Reig
Summary: This meta-analysis assessed the risk of HCC recurrence following DAA administration in HCV-related cirrhosis and HCC patients. Results showed an increased risk of recurrence and death in patients treated with DAAs, but no significant difference was observed in comparison to the DAA-unexposed group. Factors influencing recurrence included alpha-fetoprotein levels, pre-DAA recurrence history, patient performance status, and tumor burden.
Article
Gastroenterology & Hepatology
Grazia Pennisi, Marco Enea, Manuel Romero-Gomez, Mauro Vigano, Elisabetta Bugianesi, Vincent W. -s Wong, Anna Ludovica Fracanzani, Giada Sebastiani, Jerome Boursier, Annalisa Berzigotti, Mohammed Eslam, Javier Ampuero, Amine Benmassaoud, Claudia La Mantia, Yuly P. Mendoza, Jacob George, Antonio Craxi, Calogero Camma', Victor de Ledinghen, Salvatore Petta
Summary: The occurrence of extrahepatic events (EHE) in patients with non-alcoholic fatty liver disease (NAFLD) is relevant and increases according to the severity of liver fibrosis, while the risk of liver-related events (LRE) is low in patients with mild fibrosis.
ALIMENTARY PHARMACOLOGY & THERAPEUTICS
(2022)
Article
Gastroenterology & Hepatology
Grazia Pennisi, Rosaria Maria Pipitone, Daniela Cabibi, Marco Enea, Manuel Romero-Gomez, Mauro Vigano, Elisabetta Bugianesi, Vincent Wai-Sun Wong, Anna Ludovica Fracanzani, Giada Sebastiani, Annalisa Berzigotti, Francesca Di Salvo, Antonino Giulio Giannone, Claudia La Mantia, Giulia Lupo, Rossana Porcasi, Federica Vernuccio, Rossella Zito, Vito Di Marco, Calogero Camma, Antonio Craxi, Victor de Ledinghen, Stefania Grimaudo, Salvatore Petta
Summary: The presence of a cholestatic pattern in patients with NAFLD predicts a higher risk of major liver-related outcomes independently from other features of liver disease.
LIVER INTERNATIONAL
(2022)
Article
Medicine, General & Internal
Roberto Cannella, Carla Camma, Francesco Matteini, Ciro Celsa, Paolo Giuffrida, Marco Enea, Albert Comelli, Alessandro Stefano, Calogero Camma, Massimo Midiri, Roberto Lagalla, Giuseppe Brancatelli, Federica Vernuccio
Summary: This study explored the potential of radiomics on gadoxetate disodium-enhanced MRI for predicting hepatocellular carcinoma (HCC) response after transarterial embolization (TAE). The results showed that radiomics models based on gadoxetate disodium-enhanced MRI can achieve good performance in predicting the response of HCCs treated with TAE.
Article
Gastroenterology & Hepatology
Grazia Pennisi, Marco Enea, Alessandra Pandolfo, Ciro Celsa, Michela Antonucci, Carlo Ciccioli, Giuseppe Infantino, Claudia La Mantia, Stefanie Parisi, Adele Tulone, Vito Di Marco, Antonio Craxi, Calogero Camma, Salvatore Petta
Summary: This study aimed to assess the accuracy of AGILE 3+, a score based on various factors, for the diagnosis of advanced fibrosis and the prediction of liver-related events in patients with NAFLD. The results showed that AGILE 3+ had higher diagnostic performance compared to FIB-4 and LSM.
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY
(2023)
Article
Gastroenterology & Hepatology
Francesco Vizzutti, Ciro Celsa, Vincenza Calvaruso, Marco Enea, Salvatore Battaglia, Laura Turco, Marco Senzolo, Silvia Nardelli, Roberto Miraglia, Davide Roccarina, Claudia Campani, Dario Saltini, Cristian Caporali, Federica Indulti, Stefano Gitto, Alberto Zanetto, Gabriele Di Maria, Marcello Bianchini, Maddalena Pecchini, Silvia Aspite, Chiara Di Bonaventura, Michele Citone, Tomas Guasconi, Fabrizio Di Benedetto, Umberto Arena, Fabrizio Fanelli, Luigi Maruzzelli, Oliviero Riggio, Patrizia Burra, Antonio Colecchia, Erica Villa, Fabio Marra, Calogero Camma, Filippo Schepis
Summary: The aim of this study was to develop and validate a multivariable model for an accurate prediction of mortality in older adults. The results showed that creatinine and sodium levels are useful predictors for decision making in older adults. Further efforts to update the prediction model with larger sample size are warranted.
Review
Cardiac & Cardiovascular Systems
Grazia Pennisi, Ciro Celsa, Marco Enea, Marco Vaccaro, Vito Di Marco, Carlo Ciccioli, Giuseppe Infantino, Claudia La Mantia, Stefanie Parisi, Federica Vernuccio, Antonio Craxi, Calogero Camma, Salvatore Petta
Summary: This study aimed to quantify histological improvement and its risk factors in NASH patients enrolled in the placebo arms of RCTs, and to compare the effect of investigational drugs for NASH on histological outcomes. The meta-analysis on placebo arms showed NASH resolution without worsening of fibrosis in 17% of patients and at least 1 stage improvement of fibrosis without worsening of NASH in 21% of patients. Phase 3 RCTs, older age, and higher AST levels were associated with progression of liver disease. Network meta-analysis suggested that Semaglutide, Pioglitazone alone or in combination with Vitamin E, Aldafermin, Lanifibranor, and Obeticholic acid were the most effective interventions for NASH improvement and fibrosis reduction.
NUTRITION METABOLISM AND CARDIOVASCULAR DISEASES
(2022)
Letter
Gastroenterology & Hepatology
Francesco Vizzutti, Ciro Celsa, Salvatore Battaglia, Roberto Miraglia, Marco Enea, Fabio Marra, Cristian Caporali, Calogero Camma, Filippo Schepis
Letter
Gastroenterology & Hepatology
Francesco Vizzutti, Ciro Celsa, Salvatore Battaglia, Roberto Miraglia, Marco Enea, Fabio Marra, Antonio Colecchia, Calogero Camma, Filippo Schepis
Article
Oncology
Simone Lasagni, Filippo Leonardi, Alessandra Pivetti, Lorenza Di Marco, Federico Ravaioli, Matteo Serenari, Stefano Gitto, Rosina Maria Critelli, Fabiola Milosa, Adriana Romanzi, Serena Mancarella, Francesco Dituri, Mattia Riefolo, Barbara Catellani, Paolo Magistri, Dante Romagnoli, Ciro Celsa, Marco Enea, Nicola de Maria, Filippo Schepis, Antonio Colecchia, Calogero Camma, Matteo Cescon, Antonietta d'Errico, Fabrizio di Benedetto, Gianluigi Giannelli, Maria Luz Martinez-Chantar, Erica Villa
Summary: By analyzing the clinical, histopathological, and biological characteristics of patients undergoing liver transplantation, this study identified the expression of endothelial angiopoietin-2 as a powerful independent predictor of post-transplant tumor recurrence and mortality. This marker could assist in more accurate assessment and prioritization of patients before liver transplantation.
FRONTIERS IN ONCOLOGY
(2022)
Letter
Gastroenterology & Hepatology
Francesco Vizzutti, Ciro Celsa, Salvatore Battaglia, Roberto Miraglia, Marco Enea, Fabio Marra, Cristian Capporali, Calogero Camma, Filippo Schepis
Article
Gastroenterology & Hepatology
Grazia Pennisi, Rosaria Maria Pipitone, Marco Enea, Antonio De Vincentis, Salvatore Battaglia, Vito Di Marco, Vincenzo Di Martino, Federica Spatola, Federica Tavaglione, Umberto Vespasiani-Gentilucci, Rossella Zito, Stefano Romeo, Calogero Camma, Antonio Craxi, Stefania Grimaudo, Salvatore Petta
Summary: This article assessed the ability of a composite score based on clinical features, metabolic comorbidities, and genetic variants to predict the outcome of patients with nonalcoholic fatty liver disease (NAFLD). The study found that the scoring system had an adequate ability to predict the outcome of NAFLD patients and was associated with the occurrence of severe liver disease.
HEPATOLOGY COMMUNICATIONS
(2022)