EViews.com

Hello,

I would like to know if is it possible to estimate MIDAS with unrestricted VAR? Or is it possible estimate unrestricted VAR with mixed frequency data?

Thanks.

alinemouracs wrote:Hello,

I would like to know if is it possible to estimate MIDAS with unrestricted VAR? Or is it possible estimate unrestricted VAR with mixed frequency data?

Thanks.

Not yet...

Hi,

I estimated an unrestricted VAR with mixed data sample, and some results were generated. How did I do that?

Representations:

Estimation Proc:
===============================
LS(NOCONST) 1 2 3 4 5 6 7 8 9 10 L DIA\R @ DIAST7\RI DIAST5\DP MES\SE MES\EP TRIMST\VP MES\IB

VAR Model:
===============================
L = C(1,1)*L(-1) + C(1,2)*L(-2) + C(1,3)*L(-3) + C(1,4)*L(-4) + C(1,5)*L(-5) + C(1,6)*L(-6) + C(1,7)*L(-7) + C(1,8)*L(-8) + C(1,9)*L(-9) + C(1,10)*L(-10) + C(1,11)*DIA\R(-1) + C(1,12)*DIA\R(-2) + C(1,13)*DIA\R(-3) + C(1,14)*DIA\R(-4) + C(1,15)*DIA\R(-5) + C(1,16)*DIA\R(-6) + C(1,17)*DIA\R(-7) + C(1,18)*DIA\R(-8) + C(1,19)*DIA\R(-9) + C(1,20)*DIA\R(-10) + C(1,21)*DIAST7\RI + C(1,22)*DIAST5\DP + C(1,23)*MES\SE + C(1,24)*MES\EP + C(1,25)*TRIMST\VP + C(1,26)*MES\IB

DIA\R = C(2,1)*L(-1) + C(2,2)*L(-2) + C(2,3)*L(-3) + C(2,4)*L(-4) + C(2,5)*L(-5) + C(2,6)*L(-6) + C(2,7)*L(-7) + C(2,8)*L(-8) + C(2,9)*LL(-9) + C(2,10)*L(-10) + C(2,11)*DIA\R(-1) + C(2,12)*DIA\R(-2) + C(2,13)*DIA\R(-3) + C(2,14)*DIA\R(-4) + C(2,15)*DIA\R(-5) + C(2,16)*DIA\R(-6) + C(2,17)*DIA\R(-7) + C(2,18)*DIA\R(-8) + C(2,19)*DIA\R(-9) + C(2,20)*DIA\R(-10) + C(2,21)*DIAST7\RI + C(2,22)*DIAST5\DP + C(2,23)*MES\SE + C(2,24)*MES\EP + C(2,25)*TRIMST\VP + C(2,26)*MES\IB


Thanks.

It just did automatic frequency conversion of those variables. Perfectly valid method, and commonly used way of handling mixed frequency data, but it isn’t a true “mixed frequency VAR”.