2002
DOI: 10.3354/cr020271
|View full text |Cite
|
Sign up to set email alerts
|

Simplified EOFs-three alternatives to rotation

Abstract: Principal component analysis (PCA) is widely used in atmospheric science, and the resulting empirical orthogonal functions (EOFs) are often rotated to aid interpretation. In this paper 3 methods are described which provide alternatives to the standard 2-stage procedure of PCA followed by rotation. The techniques are illustrated on a small example involving sea-surface temperatures in the Mediterranean. Each method is shown to give different simplified interpretations for the major sources of variation in the d… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
15
0

Year Published

2003
2003
2021
2021

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 28 publications
(15 citation statements)
references
References 20 publications
0
15
0
Order By: Relevance
“…To identify which large-scale structures, or variability modes, are connected with TX90 and TN10 we computed the unrotated and rotated (with varimax) EOFs (Jolliffe et al 2002). The association between PC_ TX90 and PC_TN10 with the large-scale variables is measured not only temporally but also spatially.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…To identify which large-scale structures, or variability modes, are connected with TX90 and TN10 we computed the unrotated and rotated (with varimax) EOFs (Jolliffe et al 2002). The association between PC_ TX90 and PC_TN10 with the large-scale variables is measured not only temporally but also spatially.…”
Section: Methodsmentioning
confidence: 99%
“…We used Empirical Orthogonal Functions to extract the signals of the TX90 and TN10 variability (Hannachi et al 2007;Jolliffe 2002). The first mode of TX90 and TN10 accounts for a large amount of variance, 72% and 67%, respectively, with spatial patterns significant in 99% of the total area at the 95% significance level.…”
Section: Methodsmentioning
confidence: 99%
“…A detailed comparison of the new technique, SCoT, rotated PCA, and Vines' (2000) simple components is given for an example involving sea surface temperatures in Jolliffe, Uddin, and Vines (2002). Tibshirani (1996) studied the dif culties involved in the interpretation of multiple regression equations.…”
Section: A Motivating Examplementioning
confidence: 99%
“…For the rotation, the varimax criterion was used. This form of rotation is the most frequently used in climatological studies because it has been found to be entirely independent of the domain (Kaiser, 1958), and simplifies the structure of the loadings (Jolliffe et al, 2002) by maximising the variance of the squared correlation coefficients (i.e. loadings) between each rotated component with each original component (Horel, 1981).…”
Section: Pca Applicationmentioning
confidence: 99%