The RATS Software Forum

Posted: **Wed Mar 01, 2017 4:34 pm**

Dear Tom,

Could you please kindly help me understand what is the reason for the following error message: ## MAT15. Subscripts Too Large or Non-Positive
Error was evaluating entry 1753

These are my codes

Code: Select all

DATA(FORMAT=XLSX,ORG=COLUMNS,SHEET="Sheet7") 1871:01 2017:02 Index
set ret = log(index/index{1})*100
print / ret
@nbercycles(down=recession)
graph(shade=recession)
# ret

@MSRegression(switch=ch,states=2) ret
# constant 
*
compute gstart=1871:2,gend=2016:12
@MSRegInitial gstart gend
@MSRegEMGeneralSetup
do emits=1,50
   @MSRegEMStep gstart gend
   disp "Iteration" emits "Log likelihood" %logl
end do emits
*
* Polish estimates with 10 iterations of BHHH
*
compute p=%xsubmat(p,1,nstates-1,1,nstates)
nonlin(parmset=regparms) betas sigsqv
nonlin(parmset=msparms)  p
frml logl = f=%MSRegFVec(t),fpt=%MSProb(t,f),log(fpt)
@MSFilterInit
maximize(start=%(%MSRegInitVariances(),pstar=%msinit()),$
  parmset=regparms+msparms,$
  method=bhhh,iters=100) logl gstart gend
*
* Smoothed probabilities of the regimes. (The EM procedures compute
* PSMOOTH as a side effect).
*
set p1smooth = psmooth(t)(1)
graph(footer="Smoothed Probability of Regime 1",max=1.0,min=0.0)
# p1smooth
print / p1smooth

The estimates seem to be consistent with my expectation

MAXIMIZE - Estimation by BHHH
Convergence in 12 Iterations. Final criterion was 0.0000092 <= 0.0000100
Monthly Data From 1871:02 To 2016:12
Usable Observations 1751
Function Value -4694.5911

Variable Coeff Std Error T-Stat Signif
************************************************************************************
1. BETAS(1)(1) 0.77042734 0.07824661 9.84614 0.00000000
2. BETAS(2)(1) -2.15178486 0.57078399 -3.76988 0.00016333
3. SIGSQV(1) 8.23814900 0.36418361 22.62087 0.00000000
4. SIGSQV(2) 59.28603151 3.30382038 17.94469 0.00000000
5. P(1,1) 0.97394910 0.00654033 148.91426 0.00000000
6. P(1,2) 0.15828346 0.03164809 5.00136 0.00000057

I am not sure where the mistake might be for the error message. Thank you in advance

Posted: **Wed Mar 01, 2017 6:17 pm**

To find where it might be, use Edit-Show Last Error

It's probably:

set p1smooth = psmooth(t)(1)

You're using a subset of the entries, so that needs the gstart gend range.

Posted: **Wed Mar 01, 2017 10:06 pm**

TomDoan wrote:To find where it might be, use Edit-Show Last Error

It's probably:

set p1smooth = psmooth(t)(1)

You're using a subset of the entries, so that needs the gstart gend range.

Thank you for the quick reply

The RATS Software Forum

## MAT15. Subscripts Too Large or Non-Positive

## MAT15. Subscripts Too Large or Non-Positive

Re: ## MAT15. Subscripts Too Large or Non-Positive

Re: ## MAT15. Subscripts Too Large or Non-Positive