Learn R Programming

fda (version 6.2.0)

plot.cca.fd: Plot Functional Canonical Correlation Weight Functions

Description

A canonical correlation analysis produces a series of pairs of functional data objects which, when used as weighting functions, successively maximize the corresponding canonical correlation between two functional data objects. Like functional principal component weight functions, successive weight within either side fo the pair are required to be orthogonal to all previous weight functions. Consequently, each successive canonical correlation will no larger than its predecessor, and more likely substantially smaller. This function plots an object of class cca.fd that results from the use of function cca.fd. Each pair of weight functions is plotted after a left mouse click indicating that you are ready for the next plot.

Usage

# S3 method for cca.fd
plot(x, cexval = 1, ...)

Value

invisible(NULL)

Arguments

x

an object of class cca.fd produced by an invocation of function cca.fd.R.

cexval

A number used to determine label sizes in the plots.

...

other arguments for 'plot'.

Details

Produces a plot of a pair of weight functions corresponding to each canonical correlation between two functional data objects.

References

Ramsay, James O., Hooker, Giles, and Graves, Spencer (2009), Functional data analysis with R and Matlab, Springer, New York.

Ramsay, James O., and Silverman, Bernard W. (2005), Functional Data Analysis, 2nd ed., Springer, New York.

Ramsay, James O., and Silverman, Bernard W. (2002), Applied Functional Data Analysis, Springer, New York.

See Also

cca.fd, pda.fd plot.pca.fd

Examples

Run this code

#  Canonical correlation analysis of knee-hip curves
oldpar <- par(no.readonly=TRUE)
gaittime   <- (1:20)/21
gaitrange  <- c(0,1)
gaitbasis  <- create.fourier.basis(gaitrange,21)
gaitnbasis <- gaitbasis$nbasis
lambda     <- 10^(-11.5)
harmaccelLfd <- vec2Lfd(c(0, 0, (2*pi)^2, 0))
gaitfdPar <- fdPar(fd(matrix(0,gaitnbasis,1), gaitbasis), harmaccelLfd, 
                   lambda)
gaitfd    <- smooth.basis(gaittime, gait, gaitfdPar)$fd
ccafdPar  <- fdPar(gaitfd, harmaccelLfd, 1e-8)
ccafd0    <- cca.fd(gaitfd[,1], gaitfd[,2], ncan=3, ccafdPar, ccafdPar)
#  display the canonical correlations
round(ccafd0$ccacorr[1:6],3)
#  plot the unrotated canonical weight functions
plot.cca.fd(ccafd0)
#  compute a VARIMAX rotation of the canonical variables
ccafd1 <- varmx.cca.fd(ccafd0)
#  plot the rotated canonical weight functions
plot.cca.fd(ccafd1)
par(oldpar)

Run the code above in your browser using DataLab