Date/Time
Date(s) - 15/03/2024
9:30 am - 12:00 pm
Category(ies) No Categories
Joint Program in Survey Methodology (JPSM)
Session on Small Area Estimation
Co-sponsor: International Association of Survey Statisticians
Friday, March 15, 2024
3:35 – 6:00 PM US Eastern Time
9:35 AM – 12:00 PM Central European Time
Register here: (Link: https://umd.zoom.us/webinar/register/WN_Gn3inNfBSky-TiSCMLvbcA#/registration )
Speakers:
William R. Bell
Senior Mathematical Statistician for Small Area Estimation, U.S. Census Bureau.
Haoyi Chen
Coordinator, Inter-Secretariat Working Groupon Household Surveys, The United Nations.
Title: From Theory to Practice: Enhancing Capacity in Small Area Estimation for SDG monitoring.
Abstract: My presentation highlights the critical role of Small Area Estimation (SAE)techniques in advancing the monitoring of the Sustainable Development Goals(SDGs), especially in its pledge in leaving no one behind. I will also cover challenges and success stories for national statistical offices to adopt SAE and success stories; as well as the effort of the UN Statistics Division and its partners in providing training and technical support to countries.
Robert E. Fay
Senior Scientist, Westat, Inc. and Research Professor, Joint Program in Survey Methodology, University of Maryland College Park.
Title: Small Area Estimation: Personal Reflections on the Past and Future
Abstract: I will briefly introduce the Fay-Herriot model, even though it may be familiar to most in the audience, and relate the less well-known backstory for the original application. I will then skip forward in time by over three decades to describe my involvement with the Rao-Yu model, which is another area-level model. Unlike the Fay-Herriot model, the Rao-Yu model applies a simple time-series model to applications where data are available over time for slowly-changing population characteristics. In particular, I will discuss past and some ongoing work with the National Crime Victimization Survey. I will suggest that the Rao-Yu model may merit more interest than it currently receives and discuss some extensions that may further improve its applicability.
Partha Lahiri
Joint Program in Survey Methodology and Department of Mathematics, University of Maryland, College Park.
Title: A nested error regression model with high-dimensional parameter for small area estimation
Abstract: I will then discuss a relatively new idea of an extension of the well-known Battese-Harter-Fuller nested error regression model to incorporate heterogeneity in regression coefficients and variance components, data driven technique for model parameter estimation, and estimation of small area parameters.
David Newhouse
Senior Economist, The World Bank
Title: The availability of geospatial data for small area estimation
Abstract: This talk will review publicly available geospatial data, primarily derived from satellite imagery that can be used in small area estimation. Publicly available geospatial data provide accurate and geographically comprehensive measures of population density, which is correlated with many socioeconomic outcomes of interest. Geospatial data can therefore be useful for small area estimation for some socioeconomic indicators such as poverty, especially in less developed countries where recent administrative data is scarce.