EViews.com

This thread is about the Morley-Wong decomposition add-in that estimates the trend and cycle of a time series given a large information set using the Beveridge-Nelson decomposition based on a large Bayesian vector autoregression.

Hi,

I think that there is the multicollinearity problem. The correlation coefficient between incgini and consgii is 0.78.

Hi,

You are right. You already did it manually. That is only possible way to get the net spillover index. I will update it sometimes later.

You are right. It does not work for Eviews 11. I will update the add-in soon. If you have version 10 it will work.

What is the version of Eviews you use?

If you don't know the restriction matrix syntax, you can use the dialogbox instead of command line

zero restriction matrix should be like this
matrix zeromatrix = @zeros(1,5)
zeromatrix(1,2)=1
zeromatrix(1,3)=1
zeromatrix(1,5)=1

What is the version?

No. it does not affect the results. If you want to label variables for the table results, use this option.

For example:

you forget response variable. Also your threshold variable should be endogenous.

I will try to include the option which provides the accumulative IRFs. Yes, you can do this manually after saving the IRFs or write code.

Just try to estimate the FEVD with the different long horizon and check the convergence.

Hi Paolo,

Yes, the add-in was updated. It now includes the option to impose the smoothing parameter (delta). It works fine. Could you post the data and the error details?

What is the version of Eviews you use? It only works for the version 11.

1- Should all the variables be in logs before VAR estimation? It depends on your variable and VAR model. 2- Should the multi-period constraints be set as point constraints, or cumulative? For instance, imagine we have qoq GDP growth in the VAR (without the logs). We want to set GDP growth to +0.2 f...