Autoregressive Moving Average ARMA(p,q) Model
Parameters : | endog : array-like
order : iterable
exog : array-like, optional
dates : array-like of datetime, optional
freq : str, optional
|
---|
Notes
If exogenous variables are given, then the model that is fit is
\phi(L)(y_t - X_t\beta) = \theta(L)\epsilon_t
where \phi and \theta are polynomials in the lag operator, L. This is the regression model with ARMA errors, or ARMAX model. This specification is used, whether or not the model is fit using conditional sum of square or maximum-likelihood, using the method argument in statsmodels.tsa.arima_model.ARMA.fit. Therefore, for now, css and mle refer to estimation methods only. This may change for the case of the css model in future versions.
Methods
fit([order, start_params, trend, method, ...]) | Fits ARMA(p,q) model using exact maximum likelihood via Kalman filter. |
from_formula(formula, data[, subset]) | Create a Model from a formula and dataframe. |
geterrors(params) | Get the errors of the ARMA process. |
hessian(params) | Compute the Hessian at params, |
information(params) | Fisher information matrix of model |
initialize() | Initialize (possibly re-initialize) a Model instance. For |
loglike(params) | Compute the log-likelihood for ARMA(p,q) model |
loglike_css(params) | Conditional Sum of Squares likelihood function. |
loglike_kalman(params) | Compute exact loglikelihood for ARMA(p,q) model using the Kalman Filter. |
predict(params[, start, end, exog, dynamic]) | ARMA model in-sample and out-of-sample prediction |
score(params) | Compute the score function at params. |
Attributes
endog_names | |
exog_names |