DISAGGREGATE—general procedure for interpolation

[TomDoan]

You can pull out the estimate of the AR(1) coefficient as %BETA(1). Its standard error is %STDERRS(1).

[mai_villaruel]

Thank you but i need to pullout the estimate for error process ARIMA(1,1,0) - (RWAR1 is an ARIMA(1,1,0) ).

i dont think they have the same estimates with AR(1) though?

and where will I put the code for the %BETA(1) and %STDERRS(1)? will it be after the @disaggregate(tsmodel=rwar1,model=loglin,factor=4,maintain=sum,print) arm_gdpa 2000:1 2012:4 arm_gdpq_rv?

Thank you.

Mai Lin

[TomDoan]

Thanks a lot! PRINT option works. For the proportional denton model, does it really give coefficient=1? I've done for several data and coefficient is equal to 1.

Yes. Proportional Denton is described in the "Interpolation and Distribution" chapter posted in the first post in the thread. It isn't designed for general pairs of series---it's for when both series are measures of the same thing.

[TomDoan]

Thank you but i need to pullout the estimate for error process ARIMA(1,1,0) - (RWAR1 is an ARIMA(1,1,0) ).

i dont think they have the same estimates with AR(1) though?

and where will I put the code for the %BETA(1) and %STDERRS(1)? will it be after the @disaggregate(tsmodel=rwar1,model=loglin,factor=4,maintain=sum,print) arm_gdpa 2000:1 2012:4 arm_gdpq_rv?

The estimation of the ARIMA model is indirect---it's on the unobservable noise model. Since it's an ARIMA(1,1,0) model, the only free coefficient is on the AR(1) term. That's %BETA(1) with a standard error of %STDERRS(1). I'm not sure what else you want, which you can display after @DISAGGREGATE.