How important is the AR1 coefficient for WTC significance levels?

posted Oct 2, 2014, 2:51 AM by Aslak Grinsted   [ updated Oct 2, 2014, 2:51 AM ]
The definition of Wavelet coherence (WTC) effectively normalizes by the local power in time frequency space. Therefore WTC is very insensitive to the noise colour used in the null-hypothesis (see Grinsted et al. 2004). It can easily be demonstrated by example:

figure('color',[1 1 1])
X=randn(200,1);
Y=randn(200,1);
subplot(3,1,1);
orig_arcoefs=[ar1(X),ar1(Y)]
wtc(X,Y)
subplot(3,1,2);
X2=smooth(X,7);
Y2=smooth(Y,5);
smoothed_arcoefs=[ar1(X2),ar1(Y2)]
wtc(X2,Y2)
subplot(3,1,3);
wtc(X2,Y2,'ar1',[0 0]) 



The three figures are very similar.



Comments