The RATS Software Forum

In this first section, you aren't declaring the type of IMPLABEL: dec vect[strings] shocklabels varlabels compute shocklabels=||"Aggregate demand","Oil","Aggregate Supply"|| compute varlabels=||"Output","Oil Prices","Unemployment rate"|| co...

So should I try to switch my code to using panel FMOLS values instead of the Johanssen results? Would that get rid of the issue? Thanks, What you need to do first is to figure out precisely what form your model is taking. What's heterogeneous and what (if anything) is homogeneous? You might want to...

In the first part, you have the following definitions as FRML's so you can estimate: frml e = x-a10-a11*x{1}-a12*temp1{1} frml mux = a10+a11*x{1}+a12*e{1} frml muT = a20+a21*f1{1}+a22*f2{1} frml d1 = beta10^2+(beta11^2)*(e{1}-gama11)^2+(beta12^2)*temp2{1}+(alpha11^2)*f1{1}^2+(alpha12^2)*f2{1}^2 frml...

So you aren't estimating the break values, but want to input them? You don't need to use bootstrapping in that case since that would have standard asymptotics.

I am visiting this topic after a while. The codes by Todd or more generally the DLM routine in RATS use the Durbin-Koopman instead of Carter-Kohn to draw the states. I was just wondering if there are any specific advantages for doing so. They are distinct algorithms for doing the same type of draw,...

At long last, we've completed work on reproducing the results from the second half of Juselius' Cointegrated VAR Model textbook. We've also made significant revisions to the coverage of the first half of the book that we published a couple of years ago. We think the completed handbook will be a valu...

You need to use the QUIT instruction, not CLOSE, to close a file that has been opened using DEDIT. So, try using QUIT instead. If it still doesn't seem to work, email us at support@estima.com with your full contact information and we'll take a look.

Regards,
Tom Maycock

You would have to copy each of the cross sections into a series in order to do that. (Though you might find that setting this all up as a RECT[SERIES] would be easier). If you keep the same arrangement, you would need something like (for statistic (i,j)) set temp 1 bootdraws = beta(t)(i,j) then do t...

If the variables aren't cointegrated, then you can run the regression in differences. You might also have to deal with some short-term dynamics, but the first difference regression would be the place to start.

Add an explicit

source dfunit.src

before the loop.

One way to do it is to use the SCRATCH command, which is normally used to create series by number. The last parameter on SCRATCH can be used to save the number one less than that of the first new series created, so if you tell it to create zero series: scratch 0 / count The variable "count"...

Ricardo, Looking at the code, I assume this command > compute ceemetric=ceemetric+(theoretical(i,1)(h)-estimated(i,1)(h))^2/%sigma(i,1)(h) is giving you an error like this ## SX17. Missing Operator or Adjacent Operands >>>>(h))^2/%sigma(i,1)(<<<< The %SIGMA array is just a regular two-dimensional sy...

You asked a similar question about a log-linearized DSGE, and the answer is quite similar---you use DLM with TYPE=SIMULATE. You just don't have to invert the log-linearization. An example is the following

If you are new to RATS, the first place to start is with the Introduction to RATS book. Work through at least the first few examples presented there. Spending 30 to 60 minutes doing that should give you a the fundamental skills you'll need, including the ability to read in data, generate graphs, do ...

See the updated description of @ADFAUTOSELECT. It should answer your question. BTW, if that's annual data, your maximum lags are way too high. 3 or 4 is the most you should even consider.