Page 4 of 4

Re: Diebold-Yilmaz index

Posted: Thu May 07, 2020 6:13 pm
by dakila
No. it does not affect the results. If you want to label variables for the table results, use this option.

Re: Diebold-Yilmaz index

Posted: Sat May 09, 2020 8:09 pm
by Ning
Hello, when I choose "Generalized" to run the index I get an error that says "Dot, diagonal, and solve require a vector argument", but I can get result when I choose "Cholesky". I wonder what does that mean because I want to use generalized variance decomposition. Could you please help me?

Re: Diebold-Yilmaz index

Posted: Mon Jul 06, 2020 10:15 am
by Haoren
Could anybody help me? How to calculate the directional spillovers, net spillovers and net pairwise spillovers? My endogeneous variables are sp500 r_10y djubscom usdx, number of lags is 4, generalized, and I choose Roll. I choose save pairwise spillovers. It says Error 14 in encrypted program. All of those ps are NA.

Re: Diebold-Yilmaz index

Posted: Thu Jul 16, 2020 2:56 am
by Stataman
Hi, I am new in this forum.

I am working on my dissertation and I came across with this methodology (Diebold-Yilmaz index) that i found specially interesting.

Maybe it is a silly question but I would like to know if we must work with STATIONARY data (making for instance the 1ยบ difference and testing unit roots) whenever we are going to build the index. Or the calculations for the spillovers work also for the raw data, and there is no need for transformation?

Many Thanks

Kind Regards

Re: Diebold-Yilmaz index

Posted: Mon Jul 20, 2020 5:22 am
by Stataman
Hi,

I am performing the Diebold-Yilmaz spillover index using the Add-In.

The results are consistent for the TOTAL spillover idex. However, the output does not generate the Net splillover index.

One could just substract the two matrix output (To matrix - From Matrix) to obtain the NET SPILLOVER, however when I open those matrix I realize that there are not dates inside, that complicates the interpretation.

Furthermore, the matrix output does not contain the names of the endogeous variables, and it assign new variables names.

I would really apreciate your help.

Many thanks

Kind Regards

Re: Diebold-Yilmaz index

Posted: Mon Jul 20, 2020 6:02 pm
by dakila
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.