Time Series Analysis

Catalog Search > Engineering/Engineering Technology > 16.684

Note: This course is not available for the current semester.

Course No: 16.684; Last Offered: No Data;

Course Description

Review of Estimation of Stochastic processes: Estimation of mean, variance, autocovariance, autocorrelation and normalized autocovariance of discrete stochastic processes; Generation of White Noise Sequences and tests for white noise; Difference operations. Linear Stationary models: Autoregressive (AR) processes, Yule-Walker equations, partial correlation. Moving Average (MA) processes, invertibility conditions and solution for MA parameters; Relation between AR and MA processes; Autoregressive Moving Average (ARMA) processes, formulation and solution for ARMA parameters. Levinson-Durbin & related algorithms: Deterministic and Probabilistic Methods; Forward-Backward Prediction; Lattice Methods. Gram-Schmidt Orthogonalization method; Burg Algorithm. Linear nonstationary Models: Autoregressive Integrated moving Average processes, differencing to induce stationarity and determination of the order of differencing. Model Identification and Diagnostic Checking: Examples for model identification; Portmanteau test and other tests for residuals to check for white noise. Forecasting: Several methods for forecasting; Box-Jenkins forecasting functions, three types; Examples for the three types of forecasting; One-step linear predictors and confidence limits. Seasonal Time Series: Formulation of seasonal time series models and basic ideas of forecasting.

Prerequisites & Notes

  • Prerequisites:
  • Special Notes:
  • Credits: 3;

Questions About This Course?

Contact the Advising Center at 978-934-2474 or Continuing_Education@uml.edu

Use the Back button in your browser to go back to search results.