Too many ARMA terms

[Zappa_F]

Hi everybody,

I encountered an akward thing today. In my sample I have about 550 monthly observations. I wanted to estimate an ARMA(12,12), therefore I clicked "Quick" --> "Estimate equation" and then the usual window pops up which asks me to type in the dependent followed by the independent variables. Done as told, I typed in

"Y C AR(1) AR(2) AR(3) AR(4) AR(5) AR(6) AR(7) AR(8) AR(9) AR(10) AR(11) AR(12) MA(1) MA(2) MA(3) MA(4) MA(5) MA(6) MA(7) MA(8) MA(9) MA(10) MA(11) MA(12)"
Hitting enter -o my surprise- Eviews7 returned an error message saying:

"Too many ARMA terms"

When I deleted the last MA term (i.e. estimating an ARMA(12,11), also ARMA(11,12)) and hit the ok button again it worked perfectly fine.... Is there a default setting that lets me estimate at the utmost 24 coefficients???

Thx. Z

[EViews Gareth]

There is no way to override it.

[Zappa_F]

Hey Gareth,

I don't follow, to be honest. So does this mean that there is a limit of 24 coeffs that can be estimated. Maybe it's lost in translation since i#m not a native English speaker but what do you mean by "override"?

Z

[EViews Gareth]

There is a hard-coded limit of 24 ARMA terms. There is nothing you can do to change it.

[Zappa_F]

Thank you very much for your quick response, Gareth.

Cheers Z

[startz]

This may not be terribly helpful, but I've never seen anyone estimate an ARMA(12,12). Is there a reason you need so many terms?

[Zappa_F]

Well, the reason lies in the fact that I am programming the model described in Henry & Olekalns 2002, Southern Economic Journal on monthly dates. Therefore, I wanted to span lagged values of one year which necessitates the inclusion of 12 lags of AR and MA.

[startz]

Hmmm, can this be done by including lags of y rather than AR terms to get under the limit?