Article
Mathematics
Adewale Folaranmi Lukman, Jeza Allohibi, Segun Light Jegede, Emmanuel Taiwo Adewuyi, Segun Oke, Abdulmajeed Atiah Alharbi
Summary: A new penalized estimator based on the Kibria-Lukman estimator with L1-norms is proposed in this study for regularization and variable selection. Through simulations and real-life applications, it is found that the new method performs well in both low- and high-dimensional data and achieves better prediction accuracy than existing methods.
Article
Plant Sciences
Andre Froes de Borja Reis, Luiz Moro Rosso, Larry C. Purcell, Seth Naeve, Shaun N. Casteel, Peter Kovacs, Sotirios Archontoulis, Dan Davidson, Ignacio A. Ciampitti
Summary: This study investigated the environmental factors affecting nitrogen fixation in soybean and developed predictive models to assess the relationship between nitrogen fixation, crop productivity, and seed protein concentration. The study found that factors such as nitrogen fertilization, atmospheric vapor pressure deficit, and soil properties play key roles in nitrogen fixation. Predictive models showed a relative mean square error of 4.5% and an R-2 value of 0.69, indicating a strong predictive ability. The study's outcomes provide insight into improving nitrogen fixation within the sustainable intensification of soybean production.
FRONTIERS IN PLANT SCIENCE
(2021)
Article
Statistics & Probability
Hanzhong Liu, Jinzhu Jia
Summary: We study the estimation property of the Elastic Net estimator in high-dimensional linear regression models. By using a unified framework for high-dimensional analysis, we provide estimation error bounds for the Elastic Net estimator under both strict and weak sparsity. We show that, under the same conditions on the design matrix, the Elastic Net estimator achieves slightly better performance than the Lasso estimator by suitably choosing the tuning parameters.
Article
Public, Environmental & Occupational Health
Justice Moses K. Aheto, Henry Ofori Duah, Pascal Agbadi, Emmanuel Kweku Nakua
Summary: The study used machine learning techniques to identify predictors of malaria prevalence in children under-five in Ghana, finding significant predictors such as age, poverty level, severity of anemia, electricity access, and rural residence.
PREVENTIVE MEDICINE REPORTS
(2021)
Article
Operations Research & Management Science
Mostafa Rezaei, Ivor Cribben, Michele Samorani
Summary: Although researchers are interested in finding patterns in relational databases, selecting important attributes is challenging due to high correlation among attributes. A novel attribute selection procedure has been introduced to better handle highly correlated attributes in high dimensional data sets.
ANNALS OF OPERATIONS RESEARCH
(2021)
Article
Chemistry, Multidisciplinary
Mohamed Yusuf Hassan, Hasan Arman
Summary: The determination of rock tensile strength is crucial for engineering applications, with the Brazilian tensile strength test being commonly used. Regularization techniques and Keras sequential models based on TensorFlow neural networks are effective for predicting BTS.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Interdisciplinary Applications
Michael J. Wurm, Paul J. Rathouz, Bret M. Hanlon
Summary: The study proposed a new algorithm for ordinal regression that can be applied to model ordered or unordered categorical response data. This approach generalizes to a more flexible form and can shrink non-ordinal models towards their ordinal counterparts.
JOURNAL OF STATISTICAL SOFTWARE
(2021)
Article
Engineering, Multidisciplinary
Faridoon Khan, Amena Urooj, Saud Ahmed Khan, Saima K. Khosa, Sara Muhammadullah, Zahra Almaspoor
Summary: In this article, the performances of autometrics and machine learning techniques were compared under different conditions. It was found that all methods showed improved performance for large sample sizes. The methods retained all relevant variables in the presence of low and moderate multicollinearity and autocorrelation. However, for low and moderate multicollinearity, excluding AEnet, all methods kept many irrelevant predictors as well. In the presence of heteroscedasticity, all techniques often held all relevant variables but suffered from overspecification problems.
MATHEMATICAL PROBLEMS IN ENGINEERING
(2022)
Article
Mathematics
Autcha Araveeporn
Summary: The lasso and elastic net methods are commonly used techniques for parameter estimation and variable selection, with adaptive versions incorporating adaptive weights based on the power order of the estimator. This paper compares these methods in the context of high-dimensional data classification, with results indicating that the adaptive elastic net method outperforms on small dispersion compared to the adaptive lasso method.
Article
Engineering, Civil
Meghna Chakraborty, Md Shakir Mahmud, Timothy J. Gates, Subhrajit Sinha
Summary: This study uses linear regularization algorithms to analyze the factors influencing human mobility during the COVID-19 pandemic in the United States, and predicts human mobility based on these factors. The results show that multiple factors have a significant impact on daily trips.
TRANSPORTATION RESEARCH RECORD
(2023)
Article
Statistics & Probability
Mehdi Dagdoug, Camelia Goga, David Haziza
Summary: This article discusses the features and applications of model-assisted estimators and evaluates their performance in a design-based approach and a high-dimensional dataset. The results of the study demonstrate the effectiveness of these methods in estimation.
JOURNAL OF APPLIED STATISTICS
(2023)
Article
Multidisciplinary Sciences
Pathum Kossinna, Weijia Cai, Xuewen Lu, Carrie S. Shemanko, Qingrun Zhang
Summary: This article introduces a tool called SCOPE, which integrates bootstrapped least absolute shrinkage and selection operator and coexpression analysis to obtain stable results that are insensitive to variations in the data. By applying SCOPE to cancer expression datasets, core genes capturing interaction effects in crucial pan-cancer pathways related to genome instability and DNA damage response were identified, highlighting the pivotal role of CD63 as an oncogenic driver and potential therapeutic target in kidney cancer.
Article
Mathematics
Feng Hong, Lu Tian, Viswanath Devanarayan
Summary: High-dimensional data applications require the use of statistical and machine-learning algorithms to identify optimal biomarker signatures based on patient characteristics for predicting clinical outcomes in biomedical research. Regularization, particularly L-1-based regularization, is commonly used to improve prediction performance and feature selection. However, choosing the penalty parameter for regularization can be unstable and may lead to inflated predictive performance estimates. This paper proposes a Monte Carlo approach for robust regularization parameter selection and an additional cross-validation wrapper for objectively evaluating the final model's predictive performance.
Article
Energy & Fuels
Woraphon Yamaka, Rungrapee Phadkantha, Pichayakone Rakpho
Summary: This study examines the economic and energy impacts of greenhouse gas emissions on climate change in China and the USA, utilizing three machine learning models. It found that economic factors in the two countries have slightly different effects on emissions, but renewable energy production contributes to sustainable development in both nations.
Article
Multidisciplinary Sciences
Faisal Maqbool Zahid, Shahla Faisal, Christian Heumann
Summary: In high-dimensional settings, Multiple Imputation (MI) is challenging, a semi-compatible imputation model is proposed by relaxing the lasso penalty and using a ridge penalty to address instability and convergence issues. The proposed approach shows superior performance to existing MI techniques in simulation studies and real-life datasets while addressing compatibility problems.
Article
Health Care Sciences & Services
Md Hasinur Rahaman Khan, J. Ewart H. Shaw
STATISTICAL METHODS IN MEDICAL RESEARCH
(2019)
Article
Statistics & Probability
Md Mazharul Islam, Md Hasinur Rahaman Khan, Tamanna Hawlader
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2019)
Article
Multidisciplinary Sciences
Md. Belal Hossain, Sabuj Kanti Mistry, Md Mohsin, Md Hasinur Rahaman Khan
Article
Health Care Sciences & Services
M. Mazharul Islam, Faisal Ababneh, Tahmina Akter, Hasinur Rahaman Khan
EASTERN MEDITERRANEAN HEALTH JOURNAL
(2020)
Review
Infectious Diseases
Zeeba Zahra Sultana, Farhana Ul Hoque, Joseph Beyene, Md. Akhlak-Ul-Islam, Md Hasinur Rahman Khan, Shakil Ahmed, Delwer Hossain Hawlader, Ahmed Hossain
Summary: The risk of MDR-TB is higher in HIV-positive patients, particularly in South-East Asian countries. The incidence of MDR-TB varies among different age groups and countries with different income levels.
BMC INFECTIOUS DISEASES
(2021)
Correction
Infectious Diseases
Zeeba Zahra Sultana, Farhana Ul Hoque, Joseph Beyene, Md. Akhlak-Ul-Islam, Md Hasinur Rahman Khan, Shakil Ahmed, Delwer Hossain Hawlader, Ahmed Hossain
Summary: The paper has been amended and the revised version can be accessed in the original article.
BMC INFECTIOUS DISEASES
(2021)
Article
Mathematical & Computational Biology
Afsana Mimi, Md Hasinur Rahaman Khan
Summary: This article introduces a variable selection procedure for high-dimensional censored data called the MCAR method, which performs well when handling correlated data, enabling efficient estimation and variable selection.
STATISTICS IN MEDICINE
(2021)
Review
Multidisciplinary Sciences
Md. Hasanul Banna Siam, Md. Mahbub Hasan, Shazed Mohammad Tashrif, Md Hasinur Rahaman Khan, Enayetur Raheem, Mohammad Sorowar Hossain
Summary: Despite months of lockdown and public health measures, South Asian countries are still struggling to control the COVID-19 pandemic. The review in Bangladesh highlighted increased human mobility during lockdown, higher infection and death rates among males, and a significant percentage of cases among young adults. While there was a downward trend in positive test rates, the number of new daily deaths remained largely unchanged after seven months. Our findings aim to provide valuable insights for better public health practices and policies in managing infectious diseases like COVID-19 in resource-poor developing countries.
Article
Geosciences, Multidisciplinary
Sultan Mahmud, Ferdausi Mahojabin Sumana, Md Mohsin, Md Hasinur Rahaman Khan
Summary: This study focused on the climate patterns of different climate stations in Bangladesh using adaptive clustering algorithms, revealing both regional disparities and similarities. The stations were grouped into two main clusters with distinct characteristics based on important climatological factors.
Article
Statistics & Probability
Md Hasinur Rahaman Khan, Marzan Akhter
Summary: This paper presents a variable selection technique for accelerated failure time (AFT) models by extending the ranking-based variable selection (RBVS) algorithm with the weighted least square technique. Simulation studies and real data analysis demonstrate the superiority of this method in high-dimensional data analysis.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2022)
Article
Biology
Mohammad Ali, Gias U. Ahsan, Risliana Khan, Hasinur Rahman Khan, Ahmed Hossain
BMC RESEARCH NOTES
(2020)
Article
Statistics & Probability
Moza Said Al-Balushi, M. S. Ahmed, M. Mazharul Islam, Md Hasinur Rahaman Khan
JOURNAL OF STATISTICS AND MANAGEMENT SYSTEMS
(2020)
Article
Statistics & Probability
M. Mazharul Islam, Md Hasinur Rahaman Khan
JOURNAL OF RELIABILITY AND STATISTICAL STUDIES
(2019)
Article
Public, Environmental & Occupational Health
Paritosh K. Roy, Md Hasinur R. Khan, Tahmina Akter, M. Shafiqur Rahman
SPATIAL AND SPATIO-TEMPORAL EPIDEMIOLOGY
(2019)
Article
Computer Science, Artificial Intelligence
Md Hasinur Rahaman Khan, Ahmed Hossain
FRONTIERS IN ARTIFICIAL INTELLIGENCE
(2020)