A Time-Varying Multivariate Autoregressive Modeling of Econometric Time Series
A Time-Varying Multivariate Autoregressive Modeling of Econometric Time Series
RR82-07
Abstract
A time varying multivariate autoregressive modeling of econometric time series is shown. Deviations from trend data are modeled, Kozin's orthogonal Legendre polynomial time varying representation, a Householder transformation method of least squares modeling and the use of Akaike's AIC for subset selection are the key ideas in this method. Frequency domain relative power contribution computations yield an interpretation of the changing with time econometric relationships in the analysis of the U.S. hog, corn and farm wage series.
Others in Series
Working Paper
The Prediction of Time Series with Trends and Seasonalities
The Prediction of Time Series with Trends and Seasonalities
Working Paper
A Smoothness Priors Approach to the Modeling of Time Series with Tr...
A Smoothness Priors Approach to the Modeling of Time Series with Trend and Seasonality
Working Paper
On Backshift-Operator Polynomial Transformations to Stationarity fo...
On Backshift-Operator Polynomial Transformations to Stationarity for Single- and Multi-Component Time Series
Related Information
WORKING PAPER
Statistical Research Reports and Studies
Page Last Revised - October 28, 2021