PCA.trajectory: Principle component analysis of trajectory objects.

In metno/esd: Climate analysis and empirical-statistical downscaling (ESD) package for monthly and daily data

Principle component analysis of trajectory objects.

Description

Computes principal component analysis for trajectory data, e.g., storm tracks. Add some reference and details about the method. The PCA is based on svd.

Usage

## S3 method for class 'trajectory'
PCA(
  X,
  ...,
  neofs = 20,
  param = c("lon", "lat"),
  anomaly = TRUE,
  verbose = FALSE
)

Arguments

X	a 'trajectory' object
...	additional arguments
neofs	number of EOF patterns to include
param	parameters to include in principle component analysis.
anomaly	logical. If TRUE, subtract the first latitude/longitude from each trajectory.
verbose	TRUE - clutter the screen with messages

Examples

# Simple EOF for annual mean SST:
data(imilast.M03)
x <- subset(imilast.M03,is=list(lon=c(-20,20),lat=c(50,70)))
# PCA of longitude and latitude
pca <- PCA(x,param=c('lon','lat'))
plot(pca)
map(pca,projection='latlon')

# latitude only
pca <- PCA(x,param=c('lat'))
plot(pca)

metno/esd documentation built on April 24, 2024, 9:19 p.m.