Vector Error-Correction Models
Multivariate linear models including cointegrating relations and
exogenous predictor variables
Vector-error correction (VEC) models, or cointegrated VAR models, address nonstationarity in multivariate time series resulting from co-movements of multiple response series. For an example of an analysis using VEC modeling tools, see Model the United States Economy.
Apps
Econometric Modeler | Analyze and model econometric time series |
Functions
Topics
Interactive
- Analyze Time Series Data Using Econometric Modeler
Interactively visualize and analyze univariate or multivariate time series data. - Conduct Cointegration Test Using Econometric Modeler
Interactively test series for cointegration by using the Engle-Granger cointegration test and the Johansen cointegration test. - Specifying Multivariate Lag Operator Polynomials and Coefficient Constraints Interactively
Specify multivariate lag operator polynomial terms for time series model estimation using Econometric Modeler. - Estimate Vector Error-Correction Model Using Econometric Modeler
Interactively fit several vector error-correction (VEC) models to data. Then, select an estimated model and export it to the command line to generate forecasts.
Programmatic
- Model the United States Economy
Use a vector error-correction model as a linear alternative to the Smets-Wouters DSGE macroeconomic model. - Generate VEC Model Impulse Responses
Generate impulse responses from a VEC model. - VEC Model Monte Carlo Forecasts
Generate Monte Carlo and MMSE forecasts from a VEC model. - Exploit Cointegrating Relationships to Impute Missing Data
This example shows how to impute missing observations in a time series by exploiting a cointegrating relationship between the series and at least one other fully observed series.