Multiscale Geographically Weighted Regression

UCGIS Workshop - 2020

Mehak Sachdeva

Ph.D. candidate in Computational Spatial Science

Spatial Analysis Research Center

School of Geographical Sciences and Urban Planning

Arizona State University

Presented on 7/20/2020 - UCGIS Workshop Series

Contact information

Email: mehaksachdeva@asu.edu

Twitter handle: @MehakSachdeva_

Goal


The overall goal of the workshop is to introduce participants to multiscale spatial modeling and engage in a discussion of the scale of geographic processes and their measurement. Multiscale Geographically Weighted Regression (MGWR) is a cutting-edge modeling technique which integrates both nomothetic and ideographic approaches in geography and allows a more nuanced interpretation of place and space. Participants will be introduced to the field of local modeling and the opportunities it presents to measure the scale at which geographic processes operate. Participants will also complete hands-on training in the use of MGWR in research, and will be encouraged to use the provided code for their own future research.

Resources


Github book resource: https://mehak-sachdeva.github.io/MGWR_workshop_book/

Download the software here: https://sgsup.asu.edu/sparc/mgwr

Data used in the software: Direct download link

Python open-source package: https://github.com/pysal/mgwr

Acknowledgements


  • Professor Stewart Fotheringham and Professor Peter Kedron for their immense help with building the workshop material

  • MGWR Development Team: Ziqi Li, Taylor Oshan, Stewart Fotheringham, Wei Kang, Levi Wolf, Hanchen Yu, Mehak Sachdeva, and Sarah Bardin (Spatial Analysis Research Center (SPARC), Arizona State University, Tempe, USA)