Gabriel Agostini, doctoral student in information science at Cornell Tech, will give a talk, "Inferring Fine-Grained Migration Patterns across the United States," at noon on Wednesday, April 1, in 157 Hosler Building on the Penn State University Park campus.
The event is part of a spring seminar series hosted by the Initiative for Energy and Environmental Economics and Policy (EEEPI), and is free and open to the public.
About the talk
Fine-grained migration data, which record the number of people relocating from one geographic area to another, are essential for understanding a range of social, environmental and health phenomena. Migration data can be used to illuminate responses to environmental disasters and climate change, responses to economic stresses and opportunities patterns of social change, consequences of conflicts, effects of the COVID-19 pandemic, housing instability, urban-suburban migration patterns and political polarization, according to Agostini.
"Fine-grained migration data illuminate demographic, environmental and health phenomena,' Agostini said. "However, United States migration data have serious drawbacks: public data lack spatial granularity, and higher-resolution proprietary data suffer from multiple biases."
To address the limitations of existing migration data, Agostini developed a method that fuses high-resolution proprietary data with coarse Census data to create Migration Inference for GRAnular Trend Estimation (MIGRATE): 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 this talk, Agostini will discuss his data-fusion method to build MIGRATE, efforts to validate it, and outline findings and potential applications in future migration research.
"Our estimates are highly correlated with external ground-truth datasets and improve accuracy relative to raw proprietary data," Agostini said. "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.
Agostini's research leverages spatial machine learning methods and creates novel datasets to inform more equitable urban policies. He focuses on addressing challenges related to sparse and biased spatial data: specifically, how to transform coarse, crowdsourced and irregularly collected information into actionable insights for improved city resource allocation. He earned a bachelor of science degree in applied mathematics and a bachelor of arts degree in urban studies from Columbia University.
About EEEPI
Established in 2011, the Initiative for Energy and Environmental Economics and Policy operates as a University-wide initiative at Penn State with support from the Earth and Environmental Systems Institute and the Institute of Energy and the Environment. EEEPI seeks to catalyze research in energy and environmental systems economics across the University and to build a world-class group of economists with interests in interdisciplinary collaboration.
