Article
Oncology
Jaroslaw Chilimoniuk, Alicja Gosiewska, Jadwiga Slowik, Romano Weiss, P. Markus Deckert, Stefan Roediger, Michal Burdukiewicz
Summary: Analysis of DNA double-strand break count data is crucial in various fields, and choosing the appropriate distribution model is essential for accurate results.
ANNALS OF TRANSLATIONAL MEDICINE
(2021)
Article
Health Care Sciences & Services
Yin Liu, Jianghong Zhou, Zhanshou Chen, Xinyu Zhang
Summary: To improve estimation for count data with overdispersion and/or excess zeros, a novel estimation method called optimal weighting based on cross-validation is developed for the zero-inflated negative binomial model. A K-fold cross-validation technique is used to select the optimal weight vector. The proposed method enhances computational efficiency by deleting one group of observations, unlike the jackknife model averaging method. The newly developed method is proven to be asymptotically optimal. Simulation studies and empirical applications show that it outperforms three commonly used information-based model selection methods and their model averaging counterparts.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2023)
Article
Ecology
Harlan Campbell
Summary: Researchers commonly use the Poisson generalized linear model to model the association between count data and explanatory variables, but it is important to confirm the model assumptions during data fitting to avoid violating key assumptions. Existing studies suggest that model selection based on score tests may lead to substantial inflation of false positive rates when sample sizes are small.
METHODS IN ECOLOGY AND EVOLUTION
(2021)
Article
Statistics & Probability
Yasin Altinisik
Summary: Unobserved heterogeneity and excessive number of zeros are important challenges in count modeling. This study aims to merge different solutions to address these challenges, using recently developed heterogeneous distributions. The performance of the new models is evaluated through simulation studies and real-world examples.
JOURNAL OF APPLIED STATISTICS
(2023)
Article
Agriculture, Dairy & Animal Science
Y. Gevrekci, O. I. Guneri, C. Takma, A. Yesilova
Summary: This study compares different count data models for stillbirth data and finds that the negative binomial-logit hurdle regression model is the best model.
INDIAN JOURNAL OF ANIMAL RESEARCH
(2022)
Article
Biotechnology & Applied Microbiology
Peiyao Zhao, Jiaqiang Zhu, Ying Ma, Xiang Zhou
Summary: This study comprehensively analyzes 20 spatial transcriptomics datasets collected from 11 distinct technologies, characterizes the distributional properties of expression count data, and understands the statistical nature of zero values. The results show that most genes can be well modeled by either a Poisson or a negative binomial model in most spatial transcriptomics technologies. The study also identifies key sources of overdispersion and zero inflation, including gene expression heterogeneity across tissue locations and spatial distribution of cell types.
Article
Multidisciplinary Sciences
Samarendra Das, Shesh N. Rai
Summary: This article introduces a novel statistical approach for analyzing scRNA-seq data, including cell type detection, estimation of cell capture rates, and gene specific model parameters estimation. The method takes into consideration the biological processes that lead to severe dropout events in observed UMI counts of genes, and is able to perform differential expression and other downstream analyses.
Article
Statistics & Probability
Shaul K. Bar-Lev, Ad Ridder
Summary: This paper investigates the application of two classes of exponential dispersion models in fitting count data with overdispersed and zero-inflated statistics. The findings show that these models perform exceptionally well in a wide range of cases, particularly when the data exhibit significant skewness and kurtosis.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2023)
Article
Ecology
Zachary Amir, Adia Sovie, Matthew Scott Luskin
Summary: Predator-prey dynamics are fundamental to ecology, but studying interactions directly is difficult. We developed a Bayesian hierarchical model that can infer predator-prey interactions and account for differences and uncertainties. Using this model, we analyzed camera trapping datasets from tropical forest landscapes in Southeast Asia and estimated four predator-prey relationships.
ECOLOGY AND EVOLUTION
(2022)
Article
Mathematics
Lianyong Qian, Fukang Zhu
Summary: In this article, a more flexible integer-valued GARCH model based on the generalized Conway-Maxwell-Poisson distribution is proposed to model time series of counts. This model offers a unified framework to deal with overdispersed or underdispersed, zero-inflated and heavy-tailed time series of counts.
COMMUNICATIONS IN MATHEMATICS AND STATISTICS
(2023)
Article
Statistics & Probability
Yasin Altinisik, Emel Cankaya
Summary: This study proposes a new heterogeneous Poisson model with a capture of excess zeros, namely zero-inflated Poisson-Ailamujia (ZIPA) model, and a censored variant of this model is also developed. Extensive simulation study and real world applications are conducted to assess the performances of these new models.
BRAZILIAN JOURNAL OF PROBABILITY AND STATISTICS
(2022)
Article
Physics, Multidisciplinary
Xiaolei Zhang, Guohua Yan, Renjun Ma, Jiaxiu Li
Summary: This paper proposes a new approach to handle longitudinal binomial data with positive association, which can handle both random and zero number of trials and can accommodate overdispersion and zero inflation in the data.
Article
Statistics & Probability
Dianliang Deng, Xiaoqing Zhang
Summary: This paper proposes a new two-parameter probability distribution called Lindley-binomial (LB) distribution for analyzing proportional data with features such as over/under dispersion, sparseness, and zero inflation. The probabilistic properties of the distribution are derived and estimation algorithms are presented. The model is illustrated through three real-life datasets.
JOURNAL OF APPLIED STATISTICS
(2023)
Article
Computer Science, Information Systems
Brajesh Kumar Singh, Ravinder Kumar, R. Rama Kishore
Summary: This study proposed a biometric system using leading descriptor for finger knuckle biological trait image recognition and compared it with existing leading state-of-art finger knuckle print recognition. Experimental results showed that the performance of SURF, KAZE, and ORB descriptors are comparable and better than BRISK and MSER descriptors.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Biochemistry & Molecular Biology
Laura C. Paterson, Amy Festarini, Marilyne Stuart, Fawaz Ali, Christie Costello, Chad Boyer, Ronald Rogge, Norma Ybarra, John Kildea, Richard B. Richardson
Summary: Theoretical evaluations suggest that the radiation weighting factor for thermal neutrons differs from the current ICRP recommended value, which has implications for radiation protection in various scenarios.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2022)
Article
Statistics & Probability
Amani Almohaimeed, Jochen Einbeck
Summary: Random effect models and response transformation models are widely used statistical techniques. However, the methodology for simultaneously including both random effects and response transformation has been limited. This study develops a nonparametric profile maximum likelihood technique to address this issue.
STATISTICAL MODELLING
(2022)
Article
Biology
Juan S. Lopez, Monica Pujol-Canadell, Pedro Puig, Montserrat Ribas, Pablo Carrasco, Gemma Armengol, Joan F. Barquinero
Summary: This study developed a surface model for dose estimation using gamma-H2AX foci at different post-irradiation times, and validated its effectiveness. Results showed that the frequency of foci increases with dose and decreases with post-irradiation time for each dose.
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
(2022)
Article
Operations Research & Management Science
Llorenc Badiella, Pedro Puig, Carlos Lago-Penas, Marti Casals
Summary: This study analyzed the effects of red and yellow cards on the scoring rate in elite soccer. The results showed that after a player is sent off, the scoring rate of the penalized team decreases significantly, while the opponent's scoring rate increases. The impact of a red card is more or less maintained over time intervals when the player belongs to the Away team, but fades over time when the affected team is stronger. The relative difference in scoring rates is also influenced by the goal difference and the difference in booked players.
ANNALS OF OPERATIONS RESEARCH
(2023)
Article
Oncology
Chia-Jung Busch, Christian Hagel, Benjamin Becker, Agnes Oetting, Nikolaus Moeckelmann, Conrad Droste, Christina Moeller-Koop, Melanie Witt, Markus Blaurock, Sonja Loges, Kai Rothkamm, Christian Betz, Adrian Muenscher, Till S. Clauditz, Thorsten Rieckmann
Summary: This study investigated the expression of Axl and Gas6 in head and neck squamous cell carcinoma, finding that high expression of Axl indicates a need for adjuvant radiotherapy after surgery. Gas6 expression was found to be a positive predictor of survival in patients receiving radiotherapy.
Article
Biology
Jorge Ernesto Gonzalez Mesa, Bret Holladay, Manuel Higueras, Marina Di Giorgio, Joan Francesc Barquinero
Summary: This study tested the performance of different algorithms in inter-laboratory comparisons and evaluated the impact of considering a priori values. The Q/Hampel algorithm showed the best performance for estimating dose mean, while Algorithm B showed the best performance for estimating dose standard deviation. It is important to establish prior definitions of standard deviation in inter-laboratory dicentric assay comparisons.
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
(2023)
Article
Statistics & Probability
Tathagata Basu, Matthias C. M. Troffaes, Jochen Einbeck
Summary: We propose a cautious Bayesian variable selection routine and investigate the sensitivity of a hierarchical model. Our method allows us to understand the effect of prior elicitation on variable selection and is compared with other well-known methods.
SANKHYA-SERIES A-MATHEMATICAL STATISTICS AND PROBABILITY
(2023)
Article
Mathematical & Computational Biology
Christian H. Weiss, Pedro Puig, Boris Aleksandrov
Summary: This study derives the asymptotics of the Poisson and binomial Stein-type GoF statistics for general count distributions and investigates their performance and application in medical data.
BIOMETRICAL JOURNAL
(2023)
Article
Biology
Alfredo Hernandez, David Endesfelder, Jochen Einbeck, Pedro Puig, Mohamed Amine Benadjaoud, Manuel Higueras, Elizabeth Ainsbury, Gaetan Gruel, Ursula Oestreicher, Leonardo Barrios, Joan Francesc Barquinero
Summary: This article introduces a new open-source program called Biodose Tools for biological dosimetry laboratories. The program supports various mathematical processes, generates calibration curves and reports, and calculates radiation doses. Through consensus, the program helps to harmonize the calculation of uncertainties.
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
(2023)
Article
Mathematics
Shaul K. Bar-Lev, Apostolos Batsidis, Jochen Einbeck, Xu Liu, Panpan Ren
Summary: The class of natural exponential families (NEFs) with power variance functions (NEF-PVFs) is extensive and has various applications. In this study, a novel goodness-of-fit (gof) test was developed for fitting random samples to fixed members of this class, based on a characterization property of their cumulants. The asymptotic null distribution of the test statistic and an appropriate bootstrap scheme were derived. The performance of the test and its bootstrap counterpart was evaluated using a Monte Carlo study for the gamma distribution and real data examples for the modified Bessel distribution.
Article
Psychology, Developmental
Nadja Reissland, Jennifer Matthewson, Jochen Einbeck
Summary: Research suggests that infants exposed to hyperemesis gravidarum during pregnancy are more likely to have attention deficits, with some attributing this to prenatal effects of malnutrition and others to maternal mental health after birth. This study examines the impact of HG diagnosis on infant attention, independent of maternal stress, depression, anxiety, and attachment. Results show that HG-exposed infants have reduced attention abilities during play compared to controls, but this is not related to maternal mental health or attachment.
INFANT BEHAVIOR & DEVELOPMENT
(2023)
Article
Statistics & Probability
Pedro Puig, Jordi Valero, Amanda Fernandez-Fontelo
Summary: In statistical modeling, understanding the mechanisms that cause underdispersion is crucial. This article revisits several mechanisms that lead to underdispersed count distributions from new perspectives, and introduces new ones. These mechanisms include arrival processes, birth-death processes, and thinning mechanisms, which can generate underdispersed distributions with negative correlation. The article also provides example applications, including one related to Biodosimetry.
SCANDINAVIAN JOURNAL OF STATISTICS
(2023)
Article
Virology
Amani Almohaimeed, Jochen Einbeck
Summary: The COVID-19 pandemic has had a rapid global impact, affecting millions of people and causing serious health, social, and economic consequences. This study aims to analyze the trends and patterns of COVID-19 death rates in South East Asian countries through a sequential cross-sectional study, using reliable data from the years 2020, 2021, and 2022. The findings will provide valuable insights for public health policies and treatments.
Article
Physics, Fluids & Plasmas
Joan del Castillo, Pedro Puig
Summary: This paper introduces the location Pareto distribution as an extension of the power law distribution and provides a likelihood ratio test for model selection. The properties of the distribution and test are thoroughly investigated, and real data applications are presented. For large observations, the two models perform similarly, explaining why classical approaches are insensitive to differentiating between them. The recommended likelihood ratio test is simple and has high discrimination power when the complementary cumulative distribution function exhibits linearity on a log-log scale.
