Page 1 of 1

autcovariances output

PostPosted: Mon May 31, 2010 12:58 pm
by tovonony
Hello,
I have some trouble in finding the matrices of autocovariances in the global output variable "oo_" when running Alt_Ex1.mod. As it is mentionned in the dynare user guide (p.41) , "by default, dynare will return autocovariances with a lag of 5 plus the variance decomposition" but unfortunately i could not find it.

i will be gratefull for any help,
regards

Re: autcovariances output

PostPosted: Tue Jun 01, 2010 7:30 am
by jpfeifer
After running the model, you will find the variable "oo_" in the Matlab workspace. It is a structure. So if you double-click on it, you see the fields of this structure. One of them is named "autocorr". It is a 1 by 5 cell array, where each cell stores one matrix.
Alternatively, if you type "oo_.autocorr{1,number}" in the command window after running the mod-file, where number is between 1 and 5 , Matlab outputs the corresponding autocorrelation matrix in the command window.

Re: autcovariances output

PostPosted: Tue Jun 01, 2010 11:55 am
by tovonony
Jpfeifer, thanks for your response, i have seen this "oo_autocorr" in the workspace but not the "oo_gamma_y" (the matrices of autocovariances) which i am looking for, as specified in the user guide by Tommaso M.G. p.41, it is said that dynare gives it by default but i could'nt find it in the oo_ structure.
help!!!!

Re: autcovariances output

PostPosted: Tue Jun 01, 2010 2:11 pm
by jpfeifer
The user guide is somewhat out of date. The more current Dynare manual does not even mention this variable, so it is most probably deprecated. The manual states on p. 31:
The matrix of autocorrelation of the endogenous variables are made available in cell array oo_.autocorr. The element
number of the matrix in the cell array corresponds to the order of autocorrelation. The option ar specifies the number of
autocorrelation matrices available.

So this variable should be what you are looking for.

Re: autcovariances output

PostPosted: Tue Jun 01, 2010 3:16 pm
by tovonony
that i was thinking, thanks again for the information,