def xcorr_norm(self):
"""The cross-correlation between every pairwise combination time-series
in the object, where the zero lag correlation is normalized to be equal
to the correlation coefficient between the time-series
Returns
-------
TimeSeries: the time-dependent cross-correlation, with zero-lag
at time=0"""
tseries_length = self.input.data.shape[0]
t_points = self.input.data.shape[-1]
xcorr = np.zeros((tseries_length,
tseries_length,
t_points*2-1))
for i in xrange(tseries_length):
for j in xrange(i,tseries_length):
xcorr[i,j] = tsu.crosscov(
self.input.data[i],self.input.data[j],all_lags=True
)
xcorr[i,j] /= (xcorr[i,j,t_points])
xcorr[i,j] *= self.output[i,j]
idx = tsu.tril_indices(tseries_length,-1)
xcorr[idx[0],idx[1],...] = xcorr[idx[1],idx[0],...]
return ts.TimeSeries(xcorr,
sampling_interval=self.input.sampling_interval,
t0=-self.input.sampling_interval*t_points)
def xcorr(self):
"""The cross-correlation between every pairwise combination time-series
in the object. Uses np.correlation('full').
Returns
-------
TimeSeries: the time-dependent cross-correlation, with zero-lag
at time=0"""
tseries_length = self.input.data.shape[0]
t_points = self.input.data.shape[-1]
xcorr = np.zeros((tseries_length,
tseries_length,
t_points*2-1))
for i in xrange(tseries_length):
for j in xrange(i,tseries_length):
xcorr[i][j] = tsu.crosscov(
self.input.data[i],self.input.data[j],all_lags=True
)
idx = tsu.tril_indices(tseries_length,-1)
xcorr[idx[0],idx[1],...] = xcorr[idx[1],idx[0],...]
return ts.TimeSeries(xcorr,
sampling_interval=self.input.sampling_interval,
t0=-self.input.sampling_interval*t_points)
def test_crosscov():
N = 128
ar_seq1, _, _ = utils.ar_generator(N=N)
ar_seq2, _, _ = utils.ar_generator(N=N)
for all_lags in (True, False):
sxy = utils.crosscov(ar_seq1, ar_seq2, all_lags=all_lags)
sxy_ref = ref_crosscov(ar_seq1, ar_seq2, all_lags=all_lags)
err = sxy_ref - sxy
mse = np.dot(err, err) / N
npt.assert_(mse < 1e-12, "Large mean square error w.r.t. reference cross covariance")
def test_crosscov():
N = 128
ar_seq1, _, _ = utils.ar_generator(N=N)
ar_seq2, _, _ = utils.ar_generator(N=N)
for all_lags in (True, False):
sxy = utils.crosscov(ar_seq1, ar_seq2, all_lags=all_lags)
sxy_ref = ref_crosscov(ar_seq1, ar_seq2, all_lags=all_lags)
err = sxy_ref - sxy
mse = np.dot(err, err) / N
nt.assert_true(mse < 1e-12, \
'Large mean square error w.r.t. reference cross covariance')
