jaromir-benes
commented
4 years ago Hi Andrey
It's increasing because it's a combined contribution of both the initial condition AND the constant term (which is added every period). The contribution of the initial condition alone would be decreasing over time but the constant term is added every period.
Let's say you have an autoregressive equation x = 0.9*x{-1} + 1 + shock;
If you start your simulation from x=0, then the contribution of the initial cond+constant term is
1, 1.9, 2.71, 3.439, ..., 10
Note that the contributions will depend on the value of the initial condition itself - if you instead start at x=10, then the contributions will remain constant
10, 10, 10, ....
while starting above 10 will produce a declining contribution.
Best Jaromir
On Wed, Oct 23, 2019 at 11:33 AM AndreyDO notifications@github.com wrote:
Hi, guys!
I'm doing a shock decomposition exercise via the command 'simulate'. Why might the contribution of 'initial conditions & constant' term increase over time while theoretical autocorrelation for the variable that is being decomposed is below unity?
Best, Andrey Orlov
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/IRIS-Solutions-Team/IRIS-Toolbox/issues/220?email_source=notifications&email_token=AGCVKKS4S5Y5DZONAMIHMW3QQAK5TA5CNFSM4JD6RUYKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HTYIISQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGCVKKXS4TEGBUPP2FCBDYTQQAK5TANCNFSM4JD6RUYA .
AndreyDO
ripatti
nfaqs
Hi, guys!
I'm doing a shock decomposition exercise via the command 'simulate'. Why might the contribution of 'initial conditions & constant' term increase over time while theoretical autocorrelation for the variable that is being decomposed is below unity?
Best, Andrey Orlov