Cornell Tech Doctoral Student to Discuss Migration Patterns Across U.S.

Gabriel Agostini's talk will explore how fine-grained migration data can illuminate demographic, environmental and health phenomena.

Mar. 25, 2026 at 2:30am

Got story updates? Submit your updates here. ›

Gabriel Agostini, a doctoral student in information science at Cornell Tech, will give a talk titled "Inferring Fine-Grained Migration Patterns across the United States" at Penn State University on April 1. Agostini has developed a method to fuse high-resolution proprietary data with coarse Census data to create annual migration matrices capturing flows between 47.4 billion U.S. Census Block Group pairs - approximately four thousand times the spatial resolution of current public data. In his talk, Agostini will discuss his data-fusion method, efforts to validate it, and outline findings and potential applications in future migration research.

Why it matters

Understanding fine-grained migration data is essential for illuminating a range of social, environmental and health phenomena, from responses to environmental disasters and climate change to patterns of social change, housing instability, and political polarization. However, existing U.S. migration data have serious drawbacks, lacking spatial granularity or suffering from multiple biases.

The details

Agostini's method, called MIGRATE, fuses high-resolution proprietary data with coarse Census data to create annual migration matrices at an unprecedented level of spatial detail. The resulting dataset captures flows between 47.4 billion U.S. Census Block Group pairs, around four thousand times the resolution of current public data. Agostini will discuss the data-fusion process, validation efforts, and key findings, such as documenting demographic and temporal variation in homophily, upward mobility and moving distance, as well as revealing patterns like wildfire-driven out-migration that were invisible in previous data.

  • The talk will take place at noon on Wednesday, April 1, 2026.
  • The event is part of a spring seminar series hosted by the Initiative for Energy and Environmental Economics and Policy (EEEPI) at Penn State University Park.

The players

Gabriel Agostini

A doctoral student in information science at Cornell Tech who has developed a method to create high-resolution migration data for the United States.

Initiative for Energy and Environmental Economics and Policy (EEEPI)

A University-wide initiative at Penn State that seeks to catalyze research in energy and environmental systems economics and build a world-class group of economists with interests in interdisciplinary collaboration.

Got photos? Submit your photos here. ›

What they’re saying

“Fine-grained migration data illuminate demographic, environmental and health phenomena. However, United States migration data have serious drawbacks: public data lack spatial granularity, and higher-resolution proprietary data suffer from multiple biases.”

— Gabriel Agostini, Doctoral student, Cornell Tech

“Our estimates are highly correlated with external ground-truth datasets and improve accuracy relative to raw proprietary data. We use MIGRATE to analyze national and local migration patterns. Nationally, we document demographic and temporal variation in homophily, upward mobility and moving distance - for example, rising moves into top-income-quartile block groups and racial disparities in upward mobility. Locally, MIGRATE reveals patterns such as wildfire-driven out-migration that are invisible in coarser previous data.”

— Gabriel Agostini, Doctoral student, Cornell Tech

What’s next

The talk is free and open to the public, and is part of a spring seminar series hosted by the Initiative for Energy and Environmental Economics and Policy (EEEPI) at Penn State University Park.

The takeaway

Agostini's research leverages spatial machine learning methods and creates novel datasets to inform more equitable urban policies by addressing challenges related to sparse and biased spatial data, transforming coarse information into actionable insights for improved city resource allocation.