General Exam - Ameer Dharamshi
Committee: Daniela Witten (co-chair), Jon Wakefield (co-chair), Tyler McCormick, Kevin Jamieson (GSR)
Presentation: Inference after Selection and Prediction
Abstract: Modern data analysis is routinely structured as a series of interrelated tasks. Analysts explore their data to select interesting hypotheses, then aim to conduct tests using the same data. Or they fit complicated predictive models to their data, then aim to perform downstream inference with the predictions. Conducting valid inference in these settings is challenging: the sequential structure of the analyses renders traditional approaches invalid. This is particularly true in practical settings where sample sizes may be limited and data are unavoidably complex.
This dissertation proposes a suite of tools for performing inference in such contexts. In the first project, we develop the theory underpinning data thinning, a set of randomization strategies that partition the information about a target parameter contained in a dataset into…
Event interval: Single day event. Campus location: Hans Rosling Center for Population Health (HRC). Online Meeting Link: https://washington.zoom.us/j/9489943945. Campus room: HRC 375. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Academics.
Tuesday, April 29, 2025, 1:00 PM – 3:00 PM.
Biostatistics Seminar: Partial least squares regression: Highlights from what we learned in the past decade.
Speaker: Dennis Cook, PhD, Professor Emeritus, University of Minnesota
Presentation: Partial least squares regression: Highlights from what we learned in the past decade.
Abstract: Partial least squares (PLS) regression, which has been available for about 45 years, is a dimension reduction method that was designed to reduce the number of predictors p without requiring that the sample size n be large. It was developed mostly in Chemometrics, where it is now ingrained as a core method. There are perhaps hundreds of papers on PLS in the Chemometrics literature and new papers appear regularly.
And yet I think it fair to conclude that PLS regression has not been embraced by statistics communities as a core method, or even as a serviceable alternative that might sometimes be useful. Nor does there seem to be a common understanding as to why PLS regression should not be used. This seems a bit odd in view of the prevalence of PLS regression across the applied sciences.
Perhaps this is as it should be —…
Event interval: Single day event. Campus room: Virtual Seminar. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Lectures/Seminars.
Thursday, May 1, 2025, 3:30 PM – 4:30 PM.
Biostatistics Seminar: Financial Determinants and Consequences of Intergenerational Caregiving among US Households: A Quasi-Experimental Study
Speaker: Jing Li, PhD, Associate Professor of Health Economics, Associate Director, The CHOICE Institute, University of Washington
Presentation: Financial Determinants and Consequences of Intergenerational Caregiving among US Households: A Quasi-Experimental Study
Abstract: We use two decades of the nationally representative Health and Retirement Study (HRS) survey data (1996-2018) to examine financial determinants and consequences of intergenerational caregiving among U.S. households who provide care for their parents. We use an event study framework with household and year fixed effects to describe trajectories of household finances before and after caregiving. We further employ an instrumental variables design to examine the consequences of caregiving on household finances. We examine an extensive set of household-level financial outcomes, such as wealth, income, and spending, including intergenerational transfers. Overall, we do not find evidence that intergenerational caregiving negatively affects…
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: Lectures/Seminars.
Thursday, May 8, 2025, 3:30 PM – 4:30 PM.
Biostatistics Student-Invited Speaker Seminar: Statistical Inference in the Age of AI
Speaker: Emmanuel Candes, PhD, Barnum-Simons Chair in Mathematics and Statistics, Stanford University
Presentation: Statistical Inference in the Age of AI
Abstract: From proteomics to remote sensing, AI powered predictions are beginning to substitute for real data when collection of the latter is difficult, slow, or costly. We present recent work that leverage machine learning predictions both as a substitute for high-quality data and as a tool for guiding real data collection. In both cases, we achieve a significant boost in accuracy and power compared to classical methods.
Contact seminar coordinator for Zoom link.
Event interval: Single day event. Campus location: Student Union Building (HUB). Campus room: HUB 145. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Lectures/Seminars.
Thursday, May 15, 2025, 3:30 PM – 4:30 PM.
Biostatistics Seminar: Jing Ma, PhD Associate Professor Public Health Sciences Division, Fred Hutch
Speaker: Jing Ma, PhD Associate Professor Public Health Sciences Division, Fred Hutch
Presentation title and description 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: Lectures/Seminars.
Thursday, May 22, 2025, 3:30 PM – 4:30 PM.
Biostatistics Seminar: Resolving single-cell regulatory elements across species and contexts with CellWalker2
Speaker: Katie Pollard, PhD, Professor Epidemiology & Biostatistics; Director, Gladstone Institute of Data Science & Biotechnology; University of California, San Francisco
Presentation: Resolving single-cell regulatory elements across species and contexts with CellWalker2
Abstract: Tissues are comprised of cells with a wide range of similarities to each other, yet existing methods for single-cell genomics treat cell types as discrete labels. To address this gap, we developed CellWalker2, a graph diffusion-based model. With our open-source software package, hierarchically related cell types can be probabilistically matched across contexts and used to annotate cells, genomic regions, or gene sets. Additional features include estimating statistical significance and enabling gene expression and chromatin accessibility to be jointly modeled. Through simulation studies, we show that CellWalker2 performs better than existing methods in cell type annotation and mapping. We then use multiomics data from the brain…
Event interval: Single day event. Campus room: HRC 135. Accessibility Contact: Deb Nelson, nelsod6@uw.edu, 206-685-9323. Event Types: Lectures/Seminars.
Thursday, May 29, 2025, 3:30 PM – 4:30 PM.
Biostatistics Seminar: Jean Feng, PhD, Assistant Professor Epidemiology & Biostatistics, University of California, San Francisco
Speaker: Jean Feng, PhD, Assistant Professor Epidemiology & Biostatistics, University of California, San Francisco
Presentation title and description 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: Lectures/Seminars.
Thursday, June 5, 2025, 3:30 PM – 4:30 PM.
UW School of Public Health 2025 Graduation Celebration
UW School of Public Health 2025 Graduation Celebration, Where: McCaw Hall 321 Mercer St, Seattle, WA 98109
When: Monday; June 16th, 2025
Time: 11 am – 1:30 pm
Doors open at 10 am.
Students must RSVP to participate. Please check back later for RSVP form.
We hope you will join us in celebrating the tremendous accomplishments of the class of 2025.
Email SPH Student Academic Services staff at sphsas@uw.edu.
Event interval: Single day event. Accessibility Contact: SPH Student Academic Services staff at sphsas@uw.edu. Event Types: Ceremonies.
Monday, June 16, 2025, 11:00 AM – 1:30 PM.
McCaw Hall 321 Mercer St, Seattle, WA 98109.
For more info visit sph.washington.edu.