Hi
I have two questions.
1. When producing IRFs from an SVAR in EViews, the horizon in the IRF chart along the horizontal axis is 1, 2, 3, ... In the graph produced, is 1 the contemporaneous effect and then the subsequent ones are the dynamics, or is the contemporaneous effect not included in the chart?
2. Let's say I have a bivariate VAR with variables Y_{t} and X_{t}. Now in the VAR, X_{t} will be a function of the lags of both variables. However, much of the variance in X_{t} is explained by Y_{t}, and not Y_{t-1}. Hence should I include the variables in the VAR with Y_{t} leading by one time period? ie, estimate my VAR for y(1) in EViews notation. What does this mean for the IRFs, however, and how would this change the interpretation? Or is this not necessary because Cholesky allows for Y to contemporaneously affect X but not the reverse?
Thanks
SVAR, IRFs, and leading data
Moderators: EViews Gareth, EViews Moderator
Return to “Econometric Discussions”
Who is online
Users browsing this forum: No registered users and 31 guests