Events

2/5/2018 3:30pm Biostatistics Seminar: Hyunkeun (Ryan) Cho

Hyunkeun (Ryan) Cho, Ph.D.

Assistant Professor

Department of Biostatistics

College of Public Health

University of Iowa

February 5, 2018

“Various Statistical Models for Longitudinal Data with Application to a Randomized Controlled Trial”

Abstract: Longitudinal data arise frequently in biomedical and health studies, where each subject is repeatedly measured over time. For instance, a randomized controlled trial was conducted in the Washington, D.C. area from March 1997 through May 2002 of 267 women with current major depression. In this study, one of three treatments was randomly assigned to each participant, namely cognitive behavior therapy (CBT), antidepressant medication, and referral to community care. Longitudinal data have been used to compare the treatments and assess the response of each treatment on a patient over time with the goal being to find the most effective treatment. In this talk, various statistical models for longitudinal data will be introduced to address the aforementioned goal of the study: 1) Multiple linear regression model; 2) Personalized treatment effect model; 3) Initial severity-dependent model; 4) Generalized growth curve mean model; 5) Growth curve quantile model.

Individuals with disabilities are encouraged to attend all University of Iowa-sponsored events. If you are a person who requires an accommodation in order to participate in this program, please contact Ann Weberin advance at ann-weber@uiowa.edu or 319-384-1582.

Location: C217 CPHB College of Public Health Building

2/15/2018 4:00pm Distinguished Biomedical Scholar Lecture - Paul Ridker, MD, MPH

Dr. Paul Ridker, MD, MPH, will visit campus on February 15, 2018 to present a lecture on "Inflammation and Atherosclerosis: From Population Biology to the Bench to a Novel Treatment" at 4:00pm in the Prem Sahai Auditorium. Dr. Ridker is Senior Physician at Brigham and Women's Hospital and the Eugene Braunwald Professor of Medicine at Harvard Medical School.  This event is co-sponsored by the Department of Internal Medicine, Division of Cardiovascular Medicine and the Francois A. Abboud Cardiovascular Research Center. 

Faculty host: Donald Heistad, MD

 

The Distinguished Biomedical Scholar Lecture Series features top-tier scientists and physicians performing high-impact research in a variety of areas.

 

Most lectures, with few exceptions, will be held on the third Thursday of each month during the academic year. The seminars will be of broad interest and are open to the public all University faculty, staff and students. The schedule of current speakers can be found here.

 

Individuals with disabilities are encouraged to attend all University of Iowa-sponsored events. If you are a person with a disability who requires an accommodation in order to participate in this program, please contact Sonya Housholder in advance at (319) 335-8587 or sonya-housholder@uiowa.edu.

Location: 1110A Medical Education Research Facility

2/19/2018 3:30pm Biostatistics Seminar: Patrick Ten Eyck

Patrick Ten Eyck, Ph.D.

Director, Biostatistics, Epidemiology, and Research Design core

Institute for Clinical and Translational Science

University of Iowa Hospitals and Clinics

Adjunct Assistant Professor

Department of Biostatistics

College of Public Health

University of Iowa

February 19, 2018

“An Alternate Approach to Pseudo-Likelihood Model Selection in the Generalized Linear Mixed Modeling Framework”

Abstract:  This talk proposes and investigates an alternate approach to pseudo-likelihood model selection in the generalized linear mixed modeling framework. The problem with the natural approach to the computation of pseudo-likelihood model selection criteria is that the pseudo-data vary for each candidate model, leading to criteria based on fundamentally different goodness-of-fit statistics, rendering them incomparable. An alternative technique will be introduced that circumvents this problem. This new approach can be implemented using a SAS macro that obtains and applies the pseudo-data from the full model to fitting candidate models based on all possible subsets of predictor variables. A justification of the propriety of the resulting pseudo-likelihood selection criteria will be provided through an extensive study designed as a factorial experiment. The new method is then illustrated in a modeling application pertaining to bullying in public schools. The data set for the application is taken from three waves of the Iowa Youth Survey.

Location: C217 CPHB College of Public Health Building

2/26/2018 3:30pm Biostatistics Seminar: Jae-Kwang Kim

Jae-Kwang Kim, Ph.D.

Professor

Department of Statistics

Iowa State University

Ames, IA

“Data Integration for Big Data Analysis in Finite Population Inference”

Abstract:  In analyzing big data for finite population inference, it is critical to adjust for the selection bias in the big data.  Using an independent probability sampling, the selection bias of big data can be removed safely.  Such techniques can be called data integration for big data analysis.  In this talk, several methods of data integration are presented in the context of big data analysis.  Results from simulation studies are also presented.

 

 

 

Location: C217 CPHB College of Public Health Building