Principal component analysis for spatial point processes — assessing the appropriateness of the approach in an ecological context

Illian, J. , Benson, E., Crawford, J. and Staines, H. (2006) Principal component analysis for spatial point processes — assessing the appropriateness of the approach in an ecological context. In: Baddeley, A., Gregori, P., Mateu, J., Stoica, R. and Stoyan, D. (eds.) Case Studies in Spatial Point Process Modeling. Series: Lecture notes in statistics (185). Springer, pp. 135-150. ISBN 9780387283111 (doi: 10.1007/0-387-31144-0_7)

Abstract

There is a need to characterise spatial point patterns of ecological plant communities in which a very large number of points exist for many different plant species. We further investigate principal component analysis for spatial point patterns using functional data analysis tools on second-order summary statistics as introduced in [10, 11]. The approach is used to detect different types of point patterns in a multi-type pattern to classify the species by their spatial arrangement. The developed method is evaluated in a detailed feasibility study, giving rise to a number of recommendations including the choice of the appropriate summary statistic. In addition, we investigate the performance of the method under noisy conditions simulating a number of settings typically occurring in an ecological context. Overall, the method produces stable results, the best results being achieved when the pair-correlation function is used. In all settings the level of noise needs to be very high to invalidate the results.

• Item Type: Book Sections
• Additional Information: Online ISBN: 9780387311449.
• Status: Published
• Glasgow Author(s) Enlighten ID: Illian, Professor Janine
• Authors: Illian, J., Benson, E., Crawford, J., and Staines, H.
• College/School: College of Science and Engineering > School of Mathematics and Statistics > Statistics
• Publisher: Springer
• ISBN: 9780387283111