wavelet.cwt(data, dt, variance, n, pad, dj, s0, j1, lag1, param, mother)
Continuous wavelet transform from data. Wavelet params can be modified as you wish.
Parameters:
- data: array_like.
Raw of data or normalized data.
- dt: number.
Time-sample of the vector. Example: Hourly, daily, monthly, etc...
- variance: number.
Data variance.
- n: number.
Length of the data.
- pad: number/flag.
Pad the time series with zeroes to next pow of two length (recommended).
Default: pad = 1.
- dj: number.
Divide octave in sub-octaves. If dj = 0.25 this will do 4 sub-octaves per octave.
- s0: number.
The maximum frequency resolution. If it is an annual data, s0 = 2*dt say start at a scale of 6 months.
Default: s0 = 2*dt.
- j1: number.
Divide the power-of-teo with dj sub-octaves each.
Default: j1 =7/dj.
- lag1: number.
Lag-1 autocoorelation for red noise background.
Default: lag1 =0.72.
- param: number/flag.
The mother wavelet param.
Default: param = 6 (Morlet function used as default).
- mother: string.
The mother wavelet funtion.
Default: moher = ‘Morlet’.
Returns: result: dict.
Return all parameters for plot.
See also
wavelet.cwa
Notes
The Morlet wavelet is used as default int this code. The wavelet.cwt function call all lib_wavelet.py functions:
+----------------+
| cwt.py |
+----------------+
|
+----------------+
| lib_wavelet.py |
+----------------+
|
+----------------+ +----------------+
| def wavelet |--| def wave_signif|
+----------------+ +----------------+
|
+----------------+ +----------------+
| def nextpow2 |--| def wave_bases |
+----------------+ +----------------+
Example
>> dt = 0.25
>> date1 = 1871
# Test data = sst_nino3.dat is already in the package!
>> data,n,time = load_txt('sst_nino3.dat',dt,date1)
# This normalize by variance
>> data_norm, variance = normalize(data)
# Continuous wavelet transform
>> result = cwt(data_norm,0.25,variance,n,1,0.25,2*0.25,7/0.25,0.72,6,'Morlet')