def fit(self, order=(1,0,1), **kwds):
from arima import ARMA
self.mod = ARMA(self.data)
self.res = self.mod.fit(order=order, **kwds)
#self.estimated_process =
return self.res
class TsaDescriptive(object):
'''collection of descriptive statistical methods for time series
'''
def __init__(self, data, label=None, name=''):
self.data = data
self.label = label
self.name = name
def filter(self, num, den):
from scipy.signal import lfilter
xfiltered = lfilter(num, den, self.data)
return self.__class__(xfiltered, self.label, self.name + '_filtered')
def detrend(self, order=1):
import tsatools
xdetrended = tsatools.detrend(self.data, order=order)
return self.__class__(xdetrended, self.label, self.name + '_detrended')
def fit(self, order=(1,0,1), **kwds):
from arima import ARMA
self.mod = ARMA(self.data)
self.res = self.mod.fit(order=order, **kwds)
#self.estimated_process =
return self.res
def acf(self, nlags=40):
return stt.acf(self.data, nlags=nlags)
def pacf(self, nlags=40):
return stt.pacf(self.data, nlags=nlags)
def periodogram(self):
#doesn't return frequesncies
return stt.periodogram(self.data)
# copied from fftarma.py
def plot4(self, fig=None, nobs=100, nacf=20, nfreq=100):
data = self.data
acf = self.acf(nacf)
pacf = self.pacf(nacf)
w = np.linspace(0, np.pi, nfreq, endpoint=False)
spdr = self.periodogram()[:nfreq] #(w)
if fig is None:
import matplotlib.pyplot as plt
fig = plt.figure()
ax = fig.add_subplot(2,2,1)
namestr = ' for %s' % self.name if self.name else ''
ax.plot(data)
ax.set_title('Time series' + namestr)
ax = fig.add_subplot(2,2,2)
ax.plot(acf)
ax.set_title('Autocorrelation' + namestr)
ax = fig.add_subplot(2,2,3)
ax.plot(spdr) # (wr, spdr)
ax.set_title('Power Spectrum' + namestr)
ax = fig.add_subplot(2,2,4)
ax.plot(pacf)
ax.set_title('Partial Autocorrelation' + namestr)
return fig