corspec: Cross-correlation between two frequency spectra

Description

This function tests the similarity between two frequency spectra by returning their maximal correlation and the frequency shift related to it.

Usage

corspec(x, y, range, plot = TRUE, plotval = TRUE,
method = "spearman", col = "black", colval = "red",
cexval = 1, fontval = 1, xlab = "Frequency (kHz)",
ylab = "Coefficient of correlation (r)", ...)

Arguments

a first data set resulting of a spectral analysis obtained with spec or meanspec (not in dB).

a second data set resulting of a spectral analysis obtained with spec or meanspec (not in dB).

range

range of x and y (in kHz).

plot

logical, if TRUE plots r values against frequency shift (by default TRUE).

plotval

logical, if TRUE adds to the plot maximum r value and frequency offset (by default TRUE).

method

a character string indicating which correlation coefficient is to be computed ("pearson", "spearman", or "kendall") (see cor).

col

colour of r values.

colval

colour of r max and frequency offset values.

cexval

character size of r max and frequency offset values.

fontval

font of r max and frequency offset values.

xlab

title of the frequency axis.

ylab

title of the r axis.

...

other plot graphical parameters.

Value

If plot is FALSE, corspec returns a list containing four components:
rthe successive correlation values between x and y.
rmaxthe maximum correlation value between x and y.
pthe p value corresponding to rmax.
fthe frequency offset corresponding to rmax.

Details

It is important not to have data in dB. Successive correlations between x and y are computed when regularly shifting y towards lower or higher frequencies. The maximal correlation is obtained at a particular shift (frequency offset). This shift may be positive or negative. The corresponding p value, obtained with cor.test, is plotted. Inverting x and y may give slight different results, see examples.

References

Hopp, S. L., Owren, M. J. and Evans, C. S. (Eds) 1998. Animal acoustic communication. Springer, Berlin, Heidelberg.

Examples

Run this code

data(tico)
# compare the two first notes spectra
a<-spec(tico,f=22050,wl=512,at=0.2,plot=FALSE)
c<-spec(tico,f=22050,wl=512,at=1.1,plot=FALSE)
op<-par(mfrow=c(2,1), mar=c(4.5,4,3,1))
spec(tico,f=22050,wl=512,at=0.2,col="blue",type="l")
par(new=TRUE)
spec(tico,f=22050,wl=512,at=1.1,col="green",type="l")
legend(x=8,y=0.5,c("Note A", "Note C"),lty=1,col=c("blue","green"),bty="o")
par(mar=c(5,4,2,1))
corspec(a,c,range=c(0,11.025),type="l",
  ylim=c(-0.25,0.8),xaxs="i",yaxs="i",las=1)
par(op)
# different correlation methods give different results...
op<-par(mfrow=c(3,1))
corspec(a,c,range=c(0,11.025),
  type="l",xaxs="i",las=1, ylim=c(-0.25,0.8))
title("spearmann correlation (by default)")
corspec(a,c,range=c(0,11.025),
  type="l",xaxs="i",las=1,ylim=c(0,1),method="pearson")
title("pearson correlation")
corspec(a,c,range=c(0,11.025),
  type="l",xaxs="i",las=1,ylim=c(-0.23,0.5),method="kendall")
title("kendall correlation")
par(op)
# inverting x and y does not give exactly similar results
op<-par(mfrow=c(2,1),mar=c(2,4,3,1))
corspec(a,c,range=c(0,11.025),type="l")
corspec(c,a,range=c(0,11.025),type="l")
par(op)

Run the code above in your browser using DataLab

Last chance! 50% off unlimited learning