School of Public Health » Biostatistics

seminar, biostatistics

RSS XML iCal Seattle, WA — Pacific Time
This hCalendar-compliant page is optimized for search engines. View this calendar as published at www.biostat.washington.edu.

Final Exam - Katie Paulson

Presentation: Survival models for estimating child mortality rates using national-level data Abstract: In the last century, there has been substantial progress in reducing preventable child mortality globally. However, large inequalities remain both between and within countries. The United Nations Sustainable Development Goals (SDGs) include targets of fewer than 25 under-five deaths and 12 neonatal deaths per 1000 live births, in every country, by 2030. To track progress towards these SDG targets, the United Nations Inter-agency Group for Child Mortality Estimation (UN IGME) and the Global Burden of Disease (GBD) study leverage statistical models and available data to estimate child mortality rates. For countries without high-quality vital registration, data come primarily from household surveys and censuses. Currently, UN IGME and GBD both use separate modeling processes to estimate neonatal, infant, and under-five mortality rates. They also use separate modeling processes to make estimates at the national… Event interval: Single day event. Campus location: Hans Rosling Center for Population Health (HRC). Online Meeting Link: https://washington.zoom.us/j/91099894628. Campus room: HRC 370. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics. Event sponsors: UW Biostatistics. Monday, November 23, 2026, 9:00 AM – 11:00 AM.

General Exam - Yimin Zhao

Presentation: Conformal inference beyond exchangeability Abstract: We study conformal inference when the exchangeability assumption is violated. In the first project, we develop SCOPE, a semi-supervised conformal prediction framework that constructs sets of plausible terminal fates for individual cells to localize developmental windows of cell-fate specification. In multi-omic data, SCOPE also detects epigenetic priming, where chromatin-derived prediction sets resolve earlier than RNA-derived sets, and identifies candidate regulatory transcription factors. In the second project, we systematically benchmark conformal prediction methods under covariate shift using low- and high-dimensional simulations and a real-data application. We compare their target-population coverage and prediction interval widths and examine their sensitivity to modeling assumptions and tuning parameters. In the third project, we propose to investigate the theoretical properties of entropy balancing for conformal calibration, which… Event interval: Single day event. Online Meeting Link: https://washington.zoom.us/j/96524724908. Campus room: Fred Hutch, Robert M. Arnold Building, Room M3-C204, 1100 Fairview Ave. N. Seattle, WA. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics. Event sponsors: UW Biostatistics. Tuesday, November 24, 2026, 12:00 PM – 2:00 PM.

Biostatistics Seminar: Missing Data Through a Causal Lens: Toward a Complete Theory of Identification and Principled Estimation in Missing Data DAGs

Speaker: Razieh Nabi, PhD, Endowed Rollins Assistant Professor, Department of Biostatistics and Bioinformatics, Emory University Presentation: Missing Data Through a Causal Lens: Toward a Complete Theory of Identification and Principled Estimation in Missing Data DAGs Abstract: It is often said that the fundamental problem of causal inference is a missing data problem: only one potential outcome is observed for each individual. Here, we consider the converse perspective: missing data problems can be viewed as causal inference problems. Recovering the complete-data law can be interpreted as identifying the joint distribution of counterfactual values that would have been observed had all variables been measured. This connection brings tools from causal inference to bear on nonignorable missingness, while revealing important ways in which missing data identification is a distinct and, in some respects, harder problem than causal identification.  In this talk, I will discuss recent developments toward a… Event interval: Single day event. Campus location: Hans Rosling Center for Population Health (HRC). Campus room: HRC 135. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics. Event sponsors: UW Biostatistics. Thursday, December 3, 2026, 3:30 PM – 4:30 PM.

Biostatistics Seminar

Speaker, title, and abstract coming soon. Event interval: Single day event. Campus location: Hans Rosling Center for Population Health (HRC). Campus room: HRC 135. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics. Event sponsors: UW Biostatistics. Thursday, December 10, 2026, 3:30 PM – 4:30 PM.

Biostatistics Seminar: Zaid Harchaoui, PhD, Professor of Statistics, University of Washington

Location, title, and abstract TBD. Event interval: Single day event. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics. Event sponsors: UW Biostatistics. Thursday, January 7, 2027, 3:30 PM – 4:30 PM.