Warning: Cannot place an upperbound on the unbiased AR(1). Series is too short or trend is to large.

[ranzan998]

Here is the code snippet:

bm2 = detrend(bm1) cu2 = detrend(cu1) coh1,pha1,coi1,freq1,sig951 = wct(cu2,bm2,dt=1,s0=2,dj=1/8,J=-1,sig=True,significance_level=0.95,wavelet='morlet',cache=False)
For the wavelet coherence transform using above snippet, I get the following error:

/home/scilab/.virenv/py12/lib/python3.12/site-packages/pycwt/helpers.py:49: SyntaxWarning: invalid escape sequence '\g'
"""

---

Warning Traceback (most recent call last)
Cell In[3], line 30
28 bm2 = detrend(bm1)
29 cu2 = detrend(cu1)
---> 30 coh1,pha1,coi1,freq1,sig951 = wct(cu2,bm2,dt=1,s0=2,dj=1/8,J=-1,sig=True,significance_level=0.95,wavelet='morlet',cache=False)
31 #udu = (bm1 - np.mean(bm1))/np.std(bm1)
32 #goa = (cu1 - np.mean(cu1))/np.std(cu1)
33
34 #print(len(udu))

File ~/.virenv/py12/lib/python3.12/site-packages/pycwt/wavelet.py:560, in wct(y1, y2, dt, dj, s0, J, sig, significance_level, wavelet, normalize, **kwargs)
558 if sig:
559 a1, b1, c1 = ar1(y1)
--> 560 a2, b2, c2 = ar1(y2)
562 sig = wct_significance(
563 a1,
564 a2,
(...)
571 **kwargs,
572 )
573 else:

File ~/.virenv/py12/lib/python3.12/site-packages/pycwt/helpers.py:100, in ar1(x)
98 g = (-B - D0.5) / (2 * A)
99 else:
--> 100 raise Warning(
101 "Cannot place an upperbound on the unbiased AR(1). "
102 "Series is too short or trend is to large."
103 )
105 # According to Allen & Smith (1996), footnote 4
106 mu2 = -1 / N + (2 / N2) * (
107 (N - g**N) / (1 - g) - g * (1 - g ** (N - 1)) / (1 - g) ** 2
108 )

Warning: Cannot place an upperbound on the unbiased AR(1). Series is too short or trend is to large.

What could be the reason, it is detrended properly and i even tried with detrending by polyfit.