Journal of Statistics and Actuarial Research

Design and Analysis of Randomized Controlled Trials and Observational Studies, with a Focus on Addressing Sources of Bias, Confounding, and Heterogeneity in Kenya

2024-02-11T08:06:56+03:00

Purpose: The aim of the study was to investigate design and analysis of randomized controlled trials and observational studies, with a focus on addressing sources of bias, confounding, and heterogeneity

Methodology: This study adopted a desk methodology. A desk study research design is commonly known as secondary data collection. This is basically collecting data from existing resources preferably because of its low cost advantage as compared to a field research. Our current study looked into already published studies and reports as the data was easily accessed through online journals and libraries.

Findings: In Kenya, it is vital to address bias, confounding, and heterogeneity in randomized controlled trials and observational studies. Robust methodologies, including randomization and propensity score matching, are utilized to enhance validity. Consideration of local contextual factors, such as cultural norms and healthcare infrastructure, is crucial. Collaboration among researchers, policymakers, and communities is key to ensuring the quality and relevance of research for improving health outcomes in Kenya.

Unique Contribution to Theory, Practice and Policy: Randomization theory, causal inference theory & heterogeneity theory may be used to anchor future studies on design and analysis of randomized controlled trials and observational studies, with a focus on addressing sources of bias, confounding, and heterogeneity. Rigorous research methods improve the quality of evidence available to practitioners, helping them make informed decisions. High-quality research informs evidence-based policymaking.

Advancement of Statistical Theory and Methods for Survival Analysis, Longitudinal Data Analysis, and Missing Data Problems in United Kingdom

2024-02-10T00:35:01+03:00

Purpose: The aim of the study was to investigate advancement of statistical theory and methods for survival analysis, longitudinal data analysis, and missing data problems

Findings: In the United Kingdom, statistical advancements in survival analysis, longitudinal data analysis, and missing data solutions have improved predictions, enhanced the understanding of individual-level changes over time, and increased research reliability, particularly in healthcare and epidemiology. These developments have global significance, benefiting evidence-based decision-making and policy formulation.

Unique Contribution to Theory, Practice and Policy: Cox proportional-hazards model, mixed-effects models, multiple imputation may be used to anchor future studies on advancement of statistical theory and methods for survival analysis, longitudinal data analysis, and missing data problems. Create user-friendly software tools and packages for implementing advanced survival models, making them accessible to researchers and practitioners. By identifying trends and disparities over time, policymakers can design interventions and allocate resources more effectively.

Analysis of High-Dimensional and Complex Data, such as Genomic Data, Neuroimaging Data, and Text Data, Using Machine Learning and Dimension Reduction Techniques in Pakistan

2024-02-11T08:19:31+03:00

Purpose: The aim of the study was to investigate analysis of high-dimensional and complex data, such as genomic data, neuroimaging data, and text data, using machine learning and dimension reduction techniques

Findings: In Pakistan, machine learning and dimension reduction techniques have been applied to analyze high-dimensional and complex data, including genomics, neuroimaging, and text data. These efforts have led to significant advancements in disease genetics, brain imaging, and text mining. While promising, challenges such as data quality and interpretability persist, underscoring the importance of continued research and collaboration in these fields.

Unique Contribution to Theory, Practice and Policy: Social network theory, Graph theory & Complex systems theory may be used to anchor future studies on analysis of high-dimensional and complex data, such as genomic data, neuroimaging data, and text data, using machine learning and dimension reduction techniques. Apply machine learning and dimension reduction techniques to genomic data to advance the field of precision medicine. Formulate policies and regulations that address privacy and ethical concerns when dealing with sensitive data, such as genomic information and personal text data

Development and Evaluation of Statistical Models for Network Data, Such as Social Networks, Biological Networks and Brain Networks in Vietnam

2024-02-11T07:57:43+03:00

Purpose: The aim of the study was to investigate development and evaluation of statistical models for network data, such as social networks, biological networks, and brain networks

Findings: Statistical models for network data, encompassing social, biological, and brain networks, have enhanced our comprehension of these complex systems. These models reveal structural insights, dynamic patterns, and valuable applications across multiple disciplines, including epidemiology, genetics, and neuroscience. In essence, they provide indispensable tools for understanding intricate network dynamics.

Unique Contribution to Theory, Practice and Policy: Principal component analysis (PCA), Deep learning and neural networks & Information theory may be used to anchor future studies on development and evaluation of statistical models for network data, such as social networks, biological networks, and brain networks. Create user-friendly software tools and packages for implementing advanced survival models, making them accessible to researchers and practitioners. By identifying trends and disparities over time, policymakers can design interventions and allocate resources more effectively.

Application of Bayesian Methods to Causal Inference and Decision Making in Health Care and Public Policy in Uganda

2024-02-11T08:30:54+03:00

Purpose: The aim of the study was to investigate application of Bayesian methods to causal inference and decision making in health care and public policy

Findings: The use of Bayesian methods in healthcare and public policy has proven effective for estimating causal effects and making informed decisions. By integrating prior knowledge and data, Bayesian approaches enhance transparency and credibility in policy recommendations, leading to more evidence-based and effective interventions in these sectors.

Unique Contribution to Theory, Practice and Policy: Bayesian decision theory, Causal inference theory & Health economics and Bayesian analysis may be used to anchor future studies on application of Bayesian methods to causal inference and decision making in health care and public policy. Implement Bayesian methods to tailor health care interventions and treatment plans to individual patients. Encourage the integration of Bayesian methods into decision support systems for healthcare and public policy