Switching - Mixed Frequency VAR

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

alexl
Posts: 4
Joined: Sat Apr 30, 2016 6:36 am

Switching - Mixed Frequency VAR

Postby alexl » Mon May 13, 2019 12:55 am

Dear Sirs,
In Eviews 11 when I choose "VAR estimation", the VAR specification window pops-up; then I choose "Switching VAR" as the "VAR type".

In the Switching VAR, Eviews lets me put high-frequency variables and low-frequency variables (in the exogenous/endogenous variable boxes).
I also get some reasonable results.

However I cannot find, in the help, any technical details about the model used by Eviews for the estimation of this switching mixed-frequency VAR.

Is it based on Foroni, Guerin, Marcellino (2015) (https://doi.org/10.1016/j.ijforecast.2014.05.003)?

Thank you
Best regards,

Alex

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12266
Joined: Tue Sep 16, 2008 5:38 pm

Re: Switching - Mixed Frequency VAR

Postby EViews Gareth » Mon May 13, 2019 7:17 am

This post seems confused. There are two, separate, estimation types. Mixed frequency VARs and Switching VARs.

Mixed frequency VARs allow specification of high and low endogenous variables. Switching VARs do not.
Follow us on Twitter @IHSEViews

alexl
Posts: 4
Joined: Sat Apr 30, 2016 6:36 am

Re: Switching - Mixed Frequency VAR

Postby alexl » Wed May 15, 2019 2:26 pm

Dear Gareth, thank you for your prompt answer. Let me share more details on my problem.

As you said, there are two, separate, estimation types (Switching vs Mixed FRequency).
I have mixed frequency data, hence I was using the MF-VAR.

However I was thinking about 2 possible switching regimes.
Therefore I tried the Switching VAR estimation type. In the Switching VAR, eviews requires to specify an exogenous set of variables and an endogenous set of variables.
I put the low frequency variable as the exogenous variable, and the high frequency one (by writing e.g. monthly\data1xxx) as the endogenous variable.
Eviews gave me a kind of reasonable output, but I do not know which kind of estimation it did. I assumed it was a sort of application of the "Markov-Switching Mixed Frequency VAR Models", (Marcellino, Foroni, Guerin, International Journal of Forecasting 2015) but I did not find anything in the Eviews help.

I hope this can helps to clarify my doubts.
May you help me to understand which estimation Eviews does when you choose SWitching VAR and then you input variables with different frequencies?

Thank you in advance

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12266
Joined: Tue Sep 16, 2008 5:38 pm

Re: Switching - Mixed Frequency VAR

Postby EViews Gareth » Wed May 15, 2019 2:39 pm

It does not do mixed frequency switching.

When you enter variables of different frequencies in any estimator other than the specific mixed frequency estimators, EViews will do an in place frequency conversion.
Follow us on Twitter @IHSEViews


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 4 guests