Kalman Filtering and Time Series Analysis in SAS/IML

Chapter Contents

Introduction

Kalman Filtering and Time Series Analysis in SAS/IML

SAS/IML software includes a library of functions for Kalman filtering and time series analysis.

The Kalman Filter functions perform:

covariance filtering and prediction
fixed-interval smoothing
diffuse covariance filtering and prediction
diffuse fixed-interval smoothing

The time series analysis subroutines are an adaptation of parts of the TIMSAC (TIMe Series Analysis and Control) package developed by the Institute of Statistical Mathematics (ISM) in Japan.

The TIMSAC library contains subroutines to do the following:

Bayesian seasonal adjustment modeling
time series decomposition analysis
locally stationary univariate AR model fitting
locally stationary multivariate AR model fitting
multivariate AR model fittintg
periodic AR model fitting
multivariate ARMA model forecasting and forecast error variance
polynomial roots or ARMA coefficients computation
time varying coefficient AR model estimation
univariate AR model fitting

Refer to SAS/IML Software: Usage and Reference for details.

Chapter Contents
Previous
Next
Top