Date: 15:00-17:00, March 27, 2019

Venue: 東北大学 青葉山新キャンパス 環境科学研究科 本館4階講義室2

（キャンパスマップ：http://www.tohoku.ac.jp/map/ja/?f=AY_J22）

Title: Geographically Weighted Regression with flexible choices of distance metrics

Speaker: Dr. Lu Binbin, Wuhan University, China

Abstract:

Geographically Weighted Regression (GWR) has been developed as a local technique to investigation spatial nonstationarity in data relationships. GWR is calibrated with data whose influence decays with distance, distances that are commonly defined as straight line or Euclidean. However, the complexity of our real world that the scope of possible distance metrics is far larger than the trasitional Euclidean choice. In this talk, GWR with flexible choices of distance metrics will be presented, i.e. using Euclidean distance and non-Euclidean distance metrics. Furthermore, variatons in spatial relationships within a GWR model might also very in intensity with respect to location and direction. This assertion has led to extensions of GWR with parameter-specific distance metrics (PSDM GWR).