I have a model in which the dependent variable and independent variables exhibit cointegration and are non-stationary, but the model structure is an ECMX (1,0) which can be interpreted that the long-run dynamics hover near or at zero. Should a different model be used in this case or has anyone else come across this situation?
After conferring with several forecasting experts, we've concluded that this question is not clear would be better handled by contacting Technical Support. Please open a Technical Support Track and include your data and code.
Thanks.... I'm trying to resolve if this model is conceptually sound. Based on an autoregressive p= 1, the Johansen trace test results in cointegration existing between at least 2 of the 3 covariates and when I test for weak exogeneity and Granger causality tests, the model structure results in one dependendent variable (sales) and the other two being independent (gdp and hpi). The end result is a ECMX (1,0) model due to nonstationarity and cointegration with gdp and hpi explaining sales. I was informed that this model structure is conceptually unsound since there is no long run equilibrium for the variables to correct cointegration.
Registration is open! SAS is returning to Vegas for an AI and analytics experience like no other! Whether you're an executive, manager, end user or SAS partner, SAS Innovate is designed for everyone on your team. Register for just $495 by 12/31/2023.
If you are interested in speaking, there is still time to submit a session idea. More details are posted on the website.