Article
Mathematics
Sirao Wang, Jiajuan Liang, Min Zhou, Huajun Ye
Summary: The study explores testing methods for multivariate normality, proposes an improvement based on representative points, and validates the method through Monte Carlo simulations and real data analysis.
Article
Mathematics
Zhi Peng Ong, Aixiang Andy Chen, Tianming Zhu, Jin-Ting Zhang
Summary: A novel k-sample test for equal distributions for high-dimensional data is proposed in this paper, utilizing Maximum Mean Discrepancy (MMD). The test statistic is constructed using a V-statistic-based estimator of the squared MMD derived for several samples. The asymptotic null and alternative distributions of the test statistic are derived. Two simulation studies and a real data example are presented to demonstrate the effectiveness and reliability of the proposed test.
Article
Statistics & Probability
Antonello D'Ambra, Pietro Amenta, Eric J. Beh
Summary: This paper introduces an extension of correspondence analysis that takes into account ordered categorical variables, and proposes the concept of confidence region for each of the ordered categories to determine the statistical significance of a category with respect to the null hypothesis of independence. The construction of these circles for CA has not been considered in the literature, and a suitable decomposition of Taguchi's statistic is introduced to test the statistical significance of each column category.
ASTA-ADVANCES IN STATISTICAL ANALYSIS
(2021)
Article
Economics
Osman Dogan, Suleyman Taspinar, Anil K. Bera
Summary: This paper introduces a new Bayesian chi-squared test, whose asymptotic null distribution is a central chi-squared distribution. The test is robust to parametric misspecification in the alternative model and shares properties with the test based on a quadratic loss function. Four examples are provided to demonstrate the implementation of the suggested Bayesian test statistic.
JOURNAL OF ECONOMETRICS
(2021)
Article
Mathematics, Applied
Jiajuan Liang, Ping He, Jun Yang
Summary: This paper proposes a representative-point-based test for multivariate normality, which shows a considerable improvement in power compared to traditional methods according to a limited Monte Carlo study. An illustrative example demonstrates the complementary function of the new test when used together with existing methods in the literature.
Review
Biophysics
Eduardo L. Kruger, Joao Paulo Assis Gobo, Anderson Spohr Nedel, Fabio Luiz Teixeira Goncalves, Paulo Sergio Lucio, Graziela Tosini Tejas, Marina Piacenti-Silva
Summary: This systematic review provides an overview of the diverse research areas related to human biometeorology in Brazil. It highlights the research trends, main contributions, shortcomings, challenges, and prospects in this field. The review reveals a lack of international publications on the subject and identifies research gaps in certain regions of Brazil. Further research is needed in areas such as climate tourism, vector-borne diseases, and urban morbidity and mortality.
INTERNATIONAL JOURNAL OF BIOMETEOROLOGY
(2022)
Article
Computer Science, Interdisciplinary Applications
Piotr Sulewski
Summary: The paper proposes a logarithmic minimum test for three-way contingency tables and evaluates the quality of independence tests by comparing their power. Power functions are determined using the Monte Carlo method and compared, with critical values determined through simulation. The new test is shown to be more powerful than other analyzed tests, including the chi-squared test.
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION
(2021)
Article
Social Sciences, Mathematical Methods
Zawar Hussain, Salman Arif Cheema, Ishtiaq Hussain
Summary: This article discusses the correction and improvement in the Tarray, Singh, and Zaizai model, particularly when using stratified random sampling. The suggested model combines the Mangat and Singh, Mangat, and Greenberg et al. models and incorporates optional randomized response technique. Numerical results demonstrate higher efficiency compared to the Kim and Warde models as well as the aforementioned models.
SOCIOLOGICAL METHODS & RESEARCH
(2022)
Article
Mathematics, Interdisciplinary Applications
Amir Nikooienejad, Valen E. Johnson
Summary: This article introduces the concept and application of uniformly most powerful Bayesian tests (UMPBT's), proposes a derivation method, extends the scope of UMPBT's, and applies them to testing the non-centrality parameter of chi-squared distribution.
Article
Environmental Sciences
Zeying Huang, Haijun Li, Jiazhang Huang
Summary: This study employed the CHAID algorithm to explore the factors associated with consumers' adoption of nutrition facts table. Results showed that consumers' comprehension of the table, their nutrition knowledge, the usage by relatives and friends, and their focus on a healthy diet all influenced the usage.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Health Care Sciences & Services
Richard McNulty
Summary: NHST's internal logic can be analyzed using propositional calculus, with the testable H-0 determined by analyzing the range of P-values; The correspondence between H-0 and H-A must be exhaustive to avoid false dichotomies; The conclusions derived from NHST only justify that the results are not due to chance alone, rather than proving the research hypothesis is true.
BMC MEDICAL RESEARCH METHODOLOGY
(2022)
Article
Operations Research & Management Science
Gustavo L. Gilardoni, Enrico A. Colosimo, Maria Luiza G. de Toledo, Marta A. Freitas
Summary: This paper proposes a nonparametric procedure for testing the hypothesis of minimal repair and applies it to real datasets in the mining industry.
APPLIED STOCHASTIC MODELS IN BUSINESS AND INDUSTRY
(2023)
Article
Multidisciplinary Sciences
Ali A. Ismail, Lutfiah I. Al Turk
Summary: Accelerated life tests (ALTs) and partially accelerated life tests (PALTs) are effective techniques for gathering failure data rapidly without subjecting all test units to severe conditions. This article discusses frequent and Bayesian estimations of the step-stress PALTs model using time-censored data from generalized exponential distribution (GED).
JOURNAL OF KING SAUD UNIVERSITY SCIENCE
(2023)
Article
Oncology
Alexander D. Sherry, Pavlos Msaouel, Zachary R. McCaw, Joseph Abi Jaoude, Eric J. Hsu, Ramez Kouzy, Roshal Patel, Yumeng Yang, Timothy A. Lin, Cullen M. Taniguchi, Claus Roedel, Emmanouil Fokas, Chad Tang, Clifton David Fuller, Bruce Minsky, Tomer Meirson, Ryan Sun, Ethan B. Ludmir
Summary: The study reveals the presence of the 'Table 1 Fallacy' in phase III oncology trials, where significance testing is used unsoundly to compare the distributions of baseline variables between randomized groups, leading to erroneous conclusions. Factors associated with a lower likelihood of this fallacy include larger trial size, publication in European journals, and industry sponsorship.
EUROPEAN JOURNAL OF CANCER
(2023)
Article
Mathematical & Computational Biology
Michael Gaffney
Summary: In survival studies, treatment effects may not extend over the entire observation time, but shorter-term effects can still be medically important. Therefore, it is advantageous to incorporate shorter duration treatment effects into the study design. The proposed backward sequential logrank procedure allows for testing at the end of the study and sequentially at earlier time points, increasing power compared to a single test at the end of the study when the treatment effect does not extend over the entire observation period.
STATISTICS IN BIOPHARMACEUTICAL RESEARCH
(2022)
Article
Mathematical & Computational Biology
Peter C. Austin, Jiming Fang, Douglas S. Lee
Summary: The Cox proportional hazards model, commonly used in clinical and epidemiological research, assumes proportional hazards for variables. When this assumption is violated, there are two methods to allow regression coefficients to vary as a flexible function of time. This flexibility improves the modeling of data and enhances the accuracy of the model.
STATISTICS IN MEDICINE
(2022)
Article
Clinical Neurology
Salomeh Keyhani, Eric M. Cheng, Katherine Hoggatt, Peter C. Austin, Erin Madden, Paul L. Hebert, Ethan A. Halm, Ayman Naseri, Jason Johanning, Ann Abraham, Dawn M. Bravata
Summary: This study compared carotid artery stenting (CAS) to medical therapy (MT) in the treatment of asymptomatic carotid stenosis and found no difference in the risk of fatal and nonfatal stroke between the two treatment methods over a 5-year follow-up period.
Article
Cardiac & Cardiovascular Systems
Husam Abdel-Qadir, Madison Gunn, Iliana C. Lega, Andrea Pang, Peter C. Austin, Sheldon M. Singh, Cynthia A. Jackevicius, Karen Tu, Paul Dorian, Douglas S. Lee, Dennis T. Ko
Summary: The longer the duration of diabetes and the higher the glycated hemoglobin level, the significantly higher the stroke rate in patients with AF and diabetes.
JOURNAL OF THE AMERICAN HEART ASSOCIATION
(2022)
Article
Cardiac & Cardiovascular Systems
Rodolfo Rocha, Xuesong Wang, Stephen E. Fremes, Derrick Y. Tam, Dennis T. Ko, Vladimir Dzavik, Edward L. Hannan, Peter C. Austin, Maral Ouzounian, Douglas S. Lee
Summary: This study reveals the variation in the PCI:CABG ratio among hospitals and its association with clinical outcomes. Compared to low ratio institutions, high ratio institutions had a higher risk of adverse cardiac and cerebrovascular events during the follow-up period. Patients were more likely to receive PCI when the diagnostic angiogram was performed by interventional cardiologists. Having diagnostic angiogram at an institution without cardiac surgical capabilities was associated with a higher risk of adverse events.
JOURNAL OF THE AMERICAN HEART ASSOCIATION
(2022)
Article
Cardiac & Cardiovascular Systems
Husam Abdel-Qadir, Leo E. Akioyamen, Jiming Fang, Andrea Pang, Andrew C. T. Ha, Cynthia A. Jackevicius, David A. Alter, Peter C. Austin, Clare L. Atzema, R. Sacha Bhatia, Gillian L. Booth, Sharon Johnston, Irfan Dhalla, Moira K. Kapral, Harlan M. Krumholz, Candace D. McNaughton, Idan Roifman, Karen Tu, Jacob A. Udell, Harindra C. Wijeysundera, Dennis T. Ko, Michael J. Schull, Douglas S. Lee
Summary: Despite universal healthcare and prescription medication coverage in Canada, residents of socioeconomically disadvantaged neighborhoods are less likely to visit cardiologists or receive treatment after atrial fibrillation diagnosis, even though they have a higher burden of cardiovascular disease and risk of adverse outcomes.
Article
Cardiac & Cardiovascular Systems
Maneesh Sud, Anna Chu, Peter C. Austin, David J. Naimark, George Thanassoulis, Harindra C. Wijeysundera, Dennis T. Ko
Summary: By studying a large population of patients for over 10 years, we found that atherosclerotic cardiovascular disease (ASCVD) accounts for just over half of all preventable first cardiovascular events in contemporary primary prevention population. Including additional cardiovascular endpoints significantly increases the estimated incidence rates.
EUROPEAN HEART JOURNAL-QUALITY OF CARE AND CLINICAL OUTCOMES
(2023)
Article
Cardiac & Cardiovascular Systems
Joel G. Ray, Peter C. Austin, Alison L. Park, Eyal Cohen, Jiming Fang, Anna Chu
Summary: Women with a history of pre-eclampsia have a higher risk of premature coronary artery disease. The assessment of obstructive coronary artery stenosis through invasive coronary angiography has not been evaluated after pre-eclampsia.
Article
Clinical Neurology
Amy Y. X. Yu, Kara A. A. Nerenberg, Christina Diong, Jiming Fang, Anna Chu, Moira K. K. Kapral, Jodi D. D. Edwards, Sonia R. R. Dancey, Peter C. C. Austin, Nathalie Auger
Summary: This retrospective cohort study evaluated long-term maternal health outcomes after pregnancy-associated stroke and found that pregnant patients with stroke had a lower risk of death and all-cause readmission compared to non-pregnant patients with stroke at 1-year follow-up, but this difference decreased over time. Pregnant patients with stroke had a higher risk of death and readmission compared to pregnant patients without stroke at 1-year follow-up, and this association persisted for a decade.
Article
Pharmacology & Pharmacy
Guy Cafri, Peter C. Austin
Summary: Observational studies are commonly used in medicine to estimate treatment effects, and propensity score methods are often employed to minimize confounding. This article discusses methods for estimating risk difference in time-to-event outcomes using weighting or matching based on the propensity score. Monte Carlo simulations are conducted to compare the performance of these methods. The results suggest the use of weighting methods or caliper matching for point estimation and weighted robust standard errors, bootstrap methods, or matching with a naive standard error for standard error estimation.
PHARMACEUTICAL STATISTICS
(2023)
Article
Mathematical & Computational Biology
Peter C. Austin, Daniele Giardiello, Stef van Buuren
Summary: This study examined the issue of missing data when a variable is used both as an inclusion/exclusion criterion and as the primary exposure in the analysis model. Two analytic strategies were compared, and it was found that the impute-then-exclude strategy using substantive model compatible fully conditional specification had superior performance in different scenarios.
STATISTICS IN MEDICINE
(2023)
Article
Cardiac & Cardiovascular Systems
Bisan Shweikialrefaee, Dennis T. Ko, Jiming Fang, Andrea Pang, Peter C. Austin, Paul Dorian, Sheldon M. Singh, Cynthia A. Jackevicius, Karen Tu, Douglas S. Lee, Husam Abdel-Qadir
Summary: This study found that statin use is associated with a lower stroke rate in patients with atrial fibrillation (AF), while higher levels of low-density lipoprotein are associated with a higher stroke rate.
JOURNAL OF THE AMERICAN HEART ASSOCIATION
(2023)
Article
Cardiac & Cardiovascular Systems
Samuel C. C. Siu, Douglas S. S. Lee, Jiming Fang, Peter C. C. Austin, Candice K. K. Silversides
Summary: This study compared the incidence of new hypertension after pregnancy in patients with and without preexisting heart disease. It found that patients with heart disease had a higher risk of developing hypertension after pregnancy, and new hypertension was associated with subsequent death or cardiovascular events.
JOURNAL OF THE AMERICAN HEART ASSOCIATION
(2023)
Review
Cardiac & Cardiovascular Systems
Lucas C. Godoy, Dennis T. Ko, Michael E. Farkouh, Baiju R. Shah, Peter C. Austin
Summary: The Cox proportional hazards model is a popular statistical tool for modeling time to event outcomes. However, in studies comparing coronary revascularization strategies, violations of proportionality can occur due to crossing hazard functions over time. This article discusses methods for identifying and accounting for nonproportionality and provides an example case study. It also reviews recent literature on how the proportionality assumption has been reported in coronary revascularization studies.
CANADIAN JOURNAL OF CARDIOLOGY
(2023)
Article
Cardiac & Cardiovascular Systems
Hifza Buhari, Jiming Fang, Lu Han, Peter C. Austin, Paul Dorian, Cynthia A. Jackevicius, Amy Y. X. Yu, Moira K. Kapral, Sheldon M. Singh, Karen Tu, Dennis T. Ko, Clare L. Atzema, Emelia J. Benjamin, Douglas S. Lee, Husam Abdel-Qadir
Summary: This study aimed to examine the relationship between sex differences in age and cardiovascular care and stroke in atrial fibrillation (AF). The results showed that females had higher rates of stroke in AF even after adjusting for other factors. Older age and inequities in cardiovascular care partially explained the higher stroke rates in females with AF.
EUROPEAN HEART JOURNAL
(2023)
Article
Health Care Sciences & Services
Peter C. Austin, Stef van Buuren
Summary: In this study, the statistical performance of predictive mean matching and logistic regression for imputing missing binary variables was compared through Monte Carlo simulations. The results showed that the two methods had virtually identical statistical performance when the analysis model was a logistic regression model.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2023)