Nonparametric Curve Estimation: Data Exploration and Inference

The NC TraCS Biostatistics Seminar Series helps clinical and translational researchers collaborate more effectively with consulting biostatisticians by building deeper understanding of key statistical concepts and methods. Researchers then are better able to (1) evaluate relevance of the concept or method for research aim(s) definition and choice of study design; and (2) properly interpret the results of data analysis.

In this seminar, J.S. Marron, PhD, will explore smoothing methods for curve estimation that can discover useful insights in contexts where there is not an obvious statistical model, and often can also indicate poorly fitting models. Methods for understanding which discovered features represent important underlying structure will also be discussed.

Marron is the Amos Hawley Distinguished Professor in UNC's Department of Statistics and Operations Research as well as a professor in the Department of Biostatistics at the UNC Gillings School of Global Public Health. He is widely recognized as a world research leader in the statistical disciplines of high dimensional, functional and object oriented data analysis, as well as data visualization. He has made broad major contributions ranging from the invention of innovative new statistical methods, through software development and on to statistical and mathematical theory. Marron's research continues with a number of ongoing deep, interdisciplinary research collaborations with colleagues in Computer Science, Genetics, Medicine, Mathematics and Biology. A special strength is his strong record of mentoring graduate students, postdocs and junior faculty, in both statistics and also related disciplinary fields.

Register

Big Data, Big Deal? Exploring the Potential & Peril of Modern Data Sources for Social Science Research

This 5-hour course will be offered online only. Attendance is required – this course will not be recorded.

In this workshop we will offer an overview of big data types and sources and discuss their potential for survey and social science research as well as pitfalls. Specifically we will discuss social media data, digital trace data, administrative data and sensor data through a total data quality lens. We will address quality issues with each type of data, possible error frameworks to use in working with these specific data types and discuss ways to quantity data quality through a collection of metrics. We end the workshop with a discussion of how to maximize data quality for these data sources. The workshop will include several examples of work taken from the recent survey research, social science and information systems science literature and we will also include several interactive activities aimed at reinforcing key concepts around recognizing and possibly mitigating quality issues that may arise when working with these data types for social science type research.

Register

Guidelines for Design and Analysis of Stepped-Wedge Trials

This NIH Collaboratory Rethinking Clinical Trials Biostatistics Series Grand Rounds features:

Speaker: James P. Hughes, PhD
Professor Emeritus of Biostatistics, University of Washington

Moderator: Patrick J. Heagerty, PhD
Professor of Biostatistics, University of Washington

Learn More