Applying Local Indicators of Spatial Association to Analyze Longitudinal Data: The Absolute Perspective
Authors:
- 12 March 2022
- journal article
- research article
- Published by Wiley in Geographical Analysis
- Vol. 55 (2) , 225-238
Abstract
Local Indicators of Spatial Association (LISA) are a class of spatial statistical methods that have been widely applied in various scientific fields. When applying LISA to make longitudinal comparisons of spatial data, a common way is to run LISA analysis at each time point, then compare the results to infer the distributional dynamics of spatial processes. Given that LISA hinges on the global mean value that often varies across time, the LISA result generated at time Ti reflects the spatial patterns strictly with respect to Ti. Therefore, the typical comparative cross‐sectional analysis with LISA can only characterize the relative distributional dynamics. However, the relative perspective alone is inadequate to comprehend the full picture, as the patterns are not directly associated with the changes of the spatial process’s intensity. We argue that it is important to obtain the absolute distribution dynamics to complement the relative perspective, especially for tracking how spatial processes evolve across time at the local level. We develop a solution that modifies the significance test when implementing LISA analysis of longitudinal data to reveal and visualize the absolute distribution dynamics. Experiments were conducted with Mongolian livestock data and Rwanda population data.Keywords
This publication has 25 references indexed in Scilit:
- Exploratory Space–time Analysis of Inbound Tourism Flows to China CitiesInternational Journal of Tourism Research, 2013
- Light based growth indicator (LBGI): exploratory analysis of developing a proxy for local economic growth based on night lightsRegional Science Policy & Practice, 2011
- Spatial analysis of time-series changes in livestock distribution by detection of local spatial associations in MongoliaApplied Geography, 2010
- Local Indicators of Network-Constrained Clusters in Spatial Patterns Represented by a Link AttributeAnnals of the American Association of Geographers, 2010
- The Analysis of Spatial Association by Use of Distance Statistics2008
- Developing a bivariate spatial association measure: An integration of Pearson's r and Moran's IJournal of Geographical Systems, 2001
- Spatial Empirics for Economic Growth and ConvergenceGeographical Analysis, 2001
- Identifying local spatial association in flow dataJournal of Geographical Systems, 1999
- Local Spatial Autocorrelation Statistics: Distributional Issues and an ApplicationGeographical Analysis, 1995
- Local Indicators of Spatial Association—LISAGeographical Analysis, 1995