Features
Models
- Stationary time series models: AR, MA, ARMA
- Nonstationary time series models: ARIMA, SARIMA
- Univariate and multivariate time series models
- Covariance, correlation, and partial correlation functions
- Structural models: state-space form and the Kalman filter
- ARCH and GARCH models
Model Identification
- Estimation of sample covariance, correlation, and partial correlation functions
- Akaike's Information Criterion (AIC)
- Bayesian Information Criterion (BIC)
Parameter Estimation
- Yule-Walker, Levinson-Durbin, Burg's, innovations, and long autoregression algorithms
- Hannan-Rissanen procedure
- Maximum likelihood method
- Conditional maximum likelihood method
Diagnostic Checking
- Residuals
- Portmanteau, turning points, and difference-sign tests
- Information matrix
Forecasting
- Exact and approximate best linear predictors
- Updating the prediction
Spectral Analysis
- Spectra of ARMA models
- Spectrum estimation
- Smoothing spectrum
| |  | | | |
| |
|
