The RATS Software Forum

Thanks What I really want to do is: First assign value for a variable (choice), and use if-else sentence to get corresponding variables. COMPUTE choice = (1 or 2 or 3) IF choice == 1 { JJ = 1-5, 11-15 } ELSE IF choice == 2 { JJ = 1-5, 31-35 } ELSE IF choice == 3 { JJ = 16-20 } where JJ indicates the...

Hi Is that possible to run a doloop with discrete values in RATS? For example, I would like to show 1-5, 6, 7 and 11-21 instead of 1-21. I cannot figure out how to do so by the "do" instructions~~~ ******* Maybe we can create a BIG matrix and let 1-5, 6, 7, 11-21 be the element in first ro...

Dear TomDoan

I check my data file again, and I found out that the name of the data file is the reason why I can not import properly.

(I mistype a space......)

Sorry for wasting your time~~

And I really appreciate your help


Samson

hello I want to combine multiple RATS data files in one. I tried to import other data files in a selected file, but nothing happened...... (in fact, one thing did happen: series in the data files which I imported to the selected file are all disppeared!! ) Are there any limitation on the number of s...

Dear Tom, I fix the weird problem described at last post. (I forgot sourcing the msregression.src.) But the estimates of replication of Filardo(1994) are somehow different between the @msvar and the @msreg. The codes of @msreg are: compute gstart=1948:6,gend=1992:8 @MSRegression(states=2,NFIX=4, swi...

Dear Tom: You mean that the code above actually estimates a TVTP-MS model, right? But something weird occurred when I relicated Filardo(1994) with the @msregression. The codes and results from the @msregression are : Variable Coeff Std Error T-Stat Signif ********************************************...

Dear Tom, I estimated the FTP-MS model with @MSRegression you suggested previously, and it works!! (Thank you again for your kindly help :D ) Now I want to estimate the TVTP-MS model rather than FTP-MS model. It seems that @MSRegression can also estimate the TVTP-MS model according to the descriptio...

It works!!

Thank you for your kind and instant reply

Best regards,
Samson

Dear Tom: I try to estimate the MS model with the @msregression procedure. Here is the code: @MSRegression(states=2, switch = ch) r # constant shock frml logl = f=%MSRegFVec(t),fpt=%MSProb(t,f),log(fpt) nonlin(parmset=msparms) p nonlin(parmset=regparms) betas sigsqv @MSRegInitial gstart gend maximiz...

Hello, I am trying to fit a fixed-tansition-probability Markov-switching model with one exogenous variable: Y_t = b0_st + b1_st * X_t + epsilon_t , and st is either 1 or 0 . I think that the procedure MSSysRegression.src is what I need, but some errors occur when I apply this procedure. (In fact, I ...