Page 3 of 3
Re: retrieving ARIMA model of the error process
Posted: Mon Jul 14, 2014 12:22 pm
by TomDoan
You can pull out the estimate of the AR(1) coefficient as %BETA(1). Its standard error is %STDERRS(1).
Re: retrieving ARIMA model of the error process
Posted: Tue Jul 15, 2014 12:08 am
by 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
Re: DISAGGREGATE - a general procedure for interpolation
Posted: Tue Sep 22, 2015 12:20 pm
by TomDoan
Jray wrote: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.
Re: retrieving ARIMA model of the error process
Posted: Mon Jul 23, 2018 1:28 pm
by TomDoan
mai_villaruel wrote: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.