statsmodels.sandbox.tsa.fftarma.ArmaFft.impulse_response ArmaFft.impulse_response(nobs=None) get the impulse response function (MA representation) for ARMA process Parameters: ma : array_like, 1d moving average lag polynomial ar : array_like, 1d auto regressive lag polynomial nobs : int number of observations to calculate Returns: ir : array, 1d impulse response function with nobs elements Notes This is the same as finding the MA representation of an ARMA(p,q). By reversing the r

statsmodels.sandbox.tsa.fftarma.ArmaFft.generate_sample ArmaFft.generate_sample(nsample=100, scale=1.0, distrvs=None, axis=0, burnin=0) generate ARMA samples Parameters: nsample : int or tuple of ints If nsample is an integer, then this creates a 1d timeseries of length size. If nsample is a tuple, then the timeseries is along axis. All other axis have independent arma samples. scale : float standard deviation of noise distrvs : function, random number generator function that generates

statsmodels.sandbox.tsa.fftarma.ArmaFft.from_estimation classmethod ArmaFft.from_estimation(model_results, nobs=None) Create ArmaProcess instance from ARMA estimation results Parameters: model_results : ARMAResults instance A fitted model nobs : int, optional If None, nobs is taken from the results

statsmodels.sandbox.tsa.fftarma.ArmaFft.from_coeffs classmethod ArmaFft.from_coeffs(arcoefs, macoefs, nobs=100) Create ArmaProcess instance from coefficients of the lag-polynomials Parameters: arcoefs : array-like Coefficient for autoregressive lag polynomial, not including zero lag. The sign is inverted to conform to the usual time series representation of an ARMA process in statistics. See the class docstring for more information. macoefs : array-like Coefficient for moving-average lag

statsmodels.sandbox.tsa.fftarma.ArmaFft.filter2 ArmaFft.filter2(x, pad=0) [source] filter a time series using fftconvolve3 with ARMA filter padding of x currently works only if x is 1d in example it produces same observations at beginning as lfilter even without padding. TODO: this returns 1 additional observation at the end

statsmodels.sandbox.tsa.fftarma.ArmaFft.filter ArmaFft.filter(x) [source] filter a timeseries with the ARMA filter padding with zero is missing, in example I needed the padding to get initial conditions identical to direct filter Initial filtered observations differ from filter2 and signal.lfilter, but at end they are the same. See also tsa.filters.fftconvolve

statsmodels.sandbox.tsa.fftarma.ArmaFft.fftma ArmaFft.fftma(n) [source] Fourier transform of MA polynomial, zero-padded at end to n Parameters: n : int length of array after zero-padding Returns: fftar : ndarray fft of zero-padded ar polynomial

statsmodels.sandbox.tsa.fftarma.ArmaFft.fftarma ArmaFft.fftarma(n=None) [source] Fourier transform of ARMA polynomial, zero-padded at end to n The Fourier transform of the ARMA process is calculated as the ratio of the fft of the MA polynomial divided by the fft of the AR polynomial. Parameters: n : int length of array after zero-padding Returns: fftarma : ndarray fft of zero-padded arma polynomial

statsmodels.sandbox.tsa.fftarma.ArmaFft.fftar ArmaFft.fftar(n=None) [source] Fourier transform of AR polynomial, zero-padded at end to n Parameters: n : int length of array after zero-padding Returns: fftar : ndarray fft of zero-padded ar polynomial

statsmodels.sandbox.tsa.fftarma.ArmaFft.arma2ma ArmaFft.arma2ma(nobs=None)