Summary
Much statistical data is multivariate and its observations are made at different geographical locations. Analysis of such data needs to examine if and how the data’s variability depends on geography. Two approaches are examined in this paper, (1) to regress each variable onto longitudes and latitudes, (2) to compare (by regression and/or rotation) the first two principal components with longitudes and latitudes. Both approaches are complemented by examination of residuals with or without local smoothing. The paper compares the two approaches on the basis of Illinois rainfall data.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Diaconis PI, Friedman JH (1983) M and N Plots. SLAC Report PUB-2495, Stanford
Gabriel KR (1971) The Biplot—Graphical Display of Matrices with Application to Principal Component Analysis. Biometrica 58: 3453–3467
Gabriel KR (1981) Biplot, in: Kotz S, Johnson NL, Read C (eds.) Encyclopedia of Statistical Sciences. Wiley, New York, pp 23–30
Gabriel KR, Odoroff CL (1984) Resistant Lower Rank Approximations of Matrices, in: Diday E et al. (eds.) Data Analysis and Informatics. North Holland, Amsterdam, pp 23–30
Gordon AW (1981) Classification. Chapman and Hall, London
Kempton RA (1984) The Use of Biplots in Interpreting Variety by Environment Interactions. Journal of Agricultural Science 103: 123–135
Tukey PA, Tukey JW (1981) Preparation; Prechosen Sequences of Views, in: Barnett V (ed.) Interpreting Multivariate Data. Wiley, London, pp 189–213
Wegman EJ (1986) Representing Multivariate Meterological Data Using Parallel Coordinates. Preprints of the Tenth Conference on Weather Modification. American Meteorological Society, Boston
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1988 Springer-Verlag Berlin · Heidelberg
About this paper
Cite this paper
Gabriel, K.R. (1988). Some Thoughts on Comparing Multivariate Data with the Map Locations at which they are Observed. In: Gaul, W., Schader, M. (eds) Data, Expert Knowledge and Decisions. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-73489-2_29
Download citation
DOI: https://doi.org/10.1007/978-3-642-73489-2_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-73491-5
Online ISBN: 978-3-642-73489-2
eBook Packages: Springer Book Archive