The first
Computational Demography Working Group speaker of the spring quarter will be Dr. Kentaro Hoffman, and titled, “Drawing Rhinoceroses with Algorithms: Inference on Predicted Data and Its implications for Demography”. The talk is hybrid and will take place on April 1 in Raitt 223 from 10 – 11 AM PST.
Use this link to register and log onto Zoom. To receive the newsletter from
CDWG, participants may choose to join our listserv
here.
Title: “Drawing Rhinoceroses with Algorithms: Inference on Predicted Data and Its implications for Demography”
Abstract: Machine learning is increasingly used in demography to predict quantities that were once directly observed. Yet predictions are often treated as data, a practice that can lead to biased estimates and misleading uncertainty. This talk introduces Inference on Predicted Data (IPD), a framework for conducting valid statistical inference when outcomes are generated by black-box prediction models rather than measured directly.
I illustrate IPD through an application to verbal autopsies, where causes of death are inferred from free-text narratives using modern NLP methods, including large language models. While these models can achieve high predictive accuracy, naïvely using predicted causes of death in downstream analyses produces distorted demographic patterns. IPD-based corrections leverage a small amount of labeled data to recover valid estimates and uncertainty, even under prediction error and distribution shift.
The results highlight a key lesson for computational demography: accurate predictions alone are not enough for reliable population inference.
Learn more about Dr. Hoffman: Kentaro Hoffman is a statistician whose research focuses on inference with AI-generated and predicted data, uncertainty quantification, and responsible machine learning. He was previously a postdoctoral scholar at the University of Washington and Johns Hopkins University, working with Tyler McCormick, Peter Searson, and Scott Zeger. His work lies at the intersection of statistics, machine learning, and computational demography, with applications including verbal autopsies, global mortality estimation, electronic medical records, and active learning.
Request for Proposals – Interdisciplinary Network on Rural Population Health and Aging With funding from the National Institute on Aging, the Interdisciplinary Network on Rural Population Health and Aging (INRPHA) invites investigators to submit proposals for pilot research that enhances understanding of the multilevel and multidimensional drivers of rural health and aging trends and disparities. Investigators may request up to $35,000. Proposals are due by Friday, April 10.
The call for submissions for the Wittgenstein Centre Conference 2026 (WIC2026) “Demography and Human Capital” is now open!
This conference aims to advance theories, data, and multi-dimensional demographic methods for modelling human capital formation and its dynamics over time, and to connect cutting-edge evidence to policy debates globally. Human capital – education, skills, health, and capabilities – is a central driver of demographic change and a key lens for understanding social and economic development, inequality, and resilience.
We invite contributions from all disciplinary background and methodological traditions.
Key information:
Submission deadline: 30 April 2026
Conference date: Tuesday, 01 December 2026 – Wednesday, 02 December 2026
Venue: Festive Hall, Austrian Academy of Sciences, Dr. Ignaz-Seipel-Platz 2, 1010 Vienna
Format: onsite participation only
* Please note that the conference will take place immediately following the 50th anniversary celebration of the Vienna Institute of Demography (VID) on 30 November 2026 at the same venue. Conference participants are warmly invited to attend the VID’s 50th anniversary celebration as well. *