represents an ARMA process for given lag-polynomials
This is a class to bring together properties of the process. It does not do any estimation or statistical analysis.
maybe needs special handling for unit roots
Methods
acf([nobs]) | theoretical autocorrelation function of ARMA process |
acovf([nobs]) | theoretical autocovariance function of ARMA process |
ar_roots() | roots of autoregressive lag-polynomial |
arma2ar([nobs]) | |
arma2ma([nobs]) | |
from_coeffs(arcoefs, macoefs[, nobs]) | create ArmaProcess instance from coefficients of the lag-polynomials |
from_estimation(model_results[, nobs]) | create ArmaProcess instance from estimation results |
generate_sample([size, scale, distrvs, ...]) | generate ARMA samples |
impulse_response([nobs]) | get the impulse response function (MA representation) for ARMA process |
invertroots([retnew]) | make MA polynomial invertible by inverting roots inside unit circle |
isinvertible() | Arma process is invertible if MA roots are outside unit circle |
isstationary() | Arma process is stationary if AR roots are outside unit circle |
ma_roots() | roots of moving average lag-polynomial |
pacf([nobs]) | partial autocorrelation function of an ARMA process |
periodogram([nobs]) | periodogram for ARMA process given by lag-polynomials ar and ma |