scikits.statsmodels.sandbox.tsa.ARIMA.fit

ARIMA.fit(x, p, q, rhoy0=None, rhoe0=None)

estimate lag coefficients of ARMA orocess by least squares

Parameters :

x : array, 1d

time series data

p : int

number of AR lags to estimate

q : int

number of MA lags to estimate

rhoy0, rhoe0 : array_like (optional)

starting values for estimation

Returns :

rh, cov_x, infodict, mesg, ier : output of scipy.optimize.leastsq

rh :

estimate of lag parameters, concatenated [rhoy, rhoe]

cov_x :

unscaled (!) covariance matrix of coefficient estimates

Previous topic

scikits.statsmodels.sandbox.tsa.ARIMA.errfn

Next topic

scikits.statsmodels.sandbox.tsa.ARIMA.forecast

This Page