The Library
The impact of the spatial superimposition of point based statistical configurations on assessing spatial autocorrelation
Tools
Westerholt, René (2018) The impact of the spatial superimposition of point based statistical configurations on assessing spatial autocorrelation. In: AGILE 2018, Lund, Sweden, 12-15 Jun 2018. Published in: 21th AGILE Conference on Geographic Information Science pp. 1-6. ISBN 9783319782089.
|
PDF
WRAP-impact-spatial-superimposition-point-based-statistical-Westerhold-2018.pdf - Published Version - Requires a PDF viewer. Download (1119Kb) | Preview |
Official URL: https://agile-online.org/conference/proceedings/pr...
Abstract
Ambient user-generated geo-information like that from geosocial media is collected using liberal, unmoderated acquisition modes.
This offers a high degree of freedom regarding content. However, the collected information is influenced by idiosyncratic spatial perceptions. The resulting datasets are thus heterogeneous and comprise different (often inseparable), spatially and temporally superimposed statistical populations. Traditional notions of stationarity, which are oftentimes required in spatial analysis, are therefore frequently violated and conclusions about disclosed spatial structures might be misleading. This paper examines how the spatial superimposition of statistical populations influences the spatial autocorrelation estimator Moran‟s I. The approach chosen allows to gain insights beyond specific empirical datasets and with full flexibility in parameterization. A synthetic point pattern is therefore constructed, which contains two overlapping, differently scaled sub-patterns. Normally distributed values drawn from populations with different means and variances are repeatedly assigned to these, and Moran‟s I is calculated for 20,000 overall configurations. Each parameter value thereby corresponds to a multiple of the same parameter value of the other population. The results show strong influences of discrepancies in statistical parameter values of co-located populations on the characterization of spatial patterns. While differences in mean values change the magnitude of Moran‟s I, whereas differences in variances increase the range of the measure. The scale associated with the dominant of the involved populations further influences the magnitude of Moran‟s I. These results suggest that the spatial analysis of ambient user-generated geo-information from unmoderated acquisition modes may require the consideration of different superimposed statistical populations to ensure meaningful result
Item Type: | Conference Item (Paper) | ||||||
---|---|---|---|---|---|---|---|
Subjects: | H Social Sciences > HA Statistics Q Science > QA Mathematics |
||||||
Divisions: | Faculty of Social Sciences > Centre for Interdisciplinary Methodologies | ||||||
Library of Congress Subject Headings (LCSH): | Spatial analysis (Statistics) , Geospatial data | ||||||
Journal or Publication Title: | 21th AGILE Conference on Geographic Information Science | ||||||
ISBN: | 9783319782089 | ||||||
Editor: | Mansourian, A. and Pilesjö, P. and Harrie, L. and von Lammeren, R. | ||||||
Official Date: | 2018 | ||||||
Dates: |
|
||||||
Page Range: | pp. 1-6 | ||||||
Status: | Peer Reviewed | ||||||
Publication Status: | Published | ||||||
Date of first compliant deposit: | 4 March 2019 | ||||||
Date of first compliant Open Access: | 4 March 2019 | ||||||
RIOXX Funder/Project Grant: |
|
||||||
Conference Paper Type: | Paper | ||||||
Title of Event: | AGILE 2018 | ||||||
Type of Event: | Conference | ||||||
Location of Event: | Lund, Sweden | ||||||
Date(s) of Event: | 12-15 Jun 2018 |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |
Downloads
Downloads per month over past year