Related Experiment Video
Updated: May 2, 2026

Non-invasive Skeletal Muscle Quantification in Small Animals Using Micro-computed Tomography
Published on: November 8, 2024
Because Muncie's Densities Are Not Manhattan's: Using Geographical Weighting in the EM Algorithm for Areal
Jonathan P Schroeder1, David C Van Riper1
1Minnesota Population Center, University of Minnesota, Minneapolis, MN, USA.
A new geographically weighted expectation-maximization (GWEM) algorithm improves spatial data interpolation accuracy. Hybrid approaches combining GWEM with target-density weighting (TDW) offer substantial improvements for population density estimation.
Area of Science:
- Geographic Information Science
- Spatial Statistics
- Geocomputation
Background:
- Areal interpolation estimates variable distributions across zones using source data.
- Ancillary control zones often guide interpolation by estimating variable density.
- Existing methods may lack spatial variability in density estimation.
Purpose of the Study:
- Introduce a novel density estimation approach: geographically weighted expectation-maximization (GWEM).
- Evaluate GWEM's accuracy for areal interpolation using land-use/land-cover data and US census data.
- Compare GWEM with existing methods and explore hybrid approaches.
Main Methods:
- Developed the geographically weighted expectation-maximization (GWEM) algorithm, integrating expectation-maximization (EM) and geographically weighted regression.
- Applied GWEM to estimate population density using 1980 US census tract data and land-use/land-cover ancillary data.
- Compared GWEM performance against traditional methods and target-density weighting (TDW) using 1970 census data.
Main Results:
- GWEM demonstrated higher accuracy than several previously studied interpolation methods.
- Target-density weighting (TDW) using 1970 data often outperformed GWEM.
- Hybrid GWEM-TDW approaches significantly improved population density estimation accuracy.
Conclusions:
- The GWEM algorithm offers a more accurate and spatially flexible approach to density estimation in areal interpolation.
- Hybrid methods combining GWEM with TDW show strong potential for enhancing spatial data interpolation.
- Findings have implications for demographic analysis and spatial modeling requiring accurate areal interpolation.
Related Concept Videos
Gross Anatomy of Skeletal Muscles
Density and Archimedes' Principle
Apparent Weight and the Earth's Rotation
For an object on the Earth's equator, the net centripetal force that accounts for its rotation is the Earth's pull towards its center, or the weight minus the normal force that prevents it from piercing into the Earth's surface....
Muscles of the Anterior Neck
Weighted Mean
For example, consider the number of goals scored in the matches of a tournament. While computing the average number of goals scored in the tournament, it may be more important to...
Density

