Am 07.07.2020 um 13:33 schrieb Riccardo (Jack) Lucchetti:
On Tue, 7 Jul 2020, Riccardo (Jack) Lucchetti wrote:
But suppose you have a MIDAS model with an autoregressive part.
Something like
y_t = \alpha y_{t-1} + \sum_{i=0}^k \beta_i x_{\tau - i} + u_t
[...]
Does anyone have a reference on how to compute the dynamic
multipliers? I have searched a little but found no clues. Can anybody
help?
Replying to myself in case anyone is interested: I found this
... I haven't checked that one out, but I'm not sure I understand why
the problem is so difficult, at least as long as you don't have a
feedback equation for x (so no system).
Of course you have different multiplier sequences depending on the
precise hi-freq subperiod in which the shock hits. So the numbers are
going to be different for the quarterly GDP reaction when the shock hits
in February compared to, say, in March. So I'd say any function that
computes this will have to return a 2D-matrix instead of a vector
(sequence).
cheers
sven
_______________________________________________
Gretl-users mailing list -- gretl-users@gretlml.univpm.it
To unsubscribe send an email to gretl-users-le...@gretlml.univpm.it
Website:
https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/
