ARIMA Forecasting

[mtaimoor23]

Hello Everyone,I have a question which may be very simple .For Example I have estimated ARIMA(1,1,4) from a data having sample 1961-1999.To produce forecast I re-estimate sample from 1961-1995 and then its easy to produce forecast on eviews.I want to confirm forecast produced by EViews manually but I am not able to do so.Anyone can guide me.

[EViews Gareth]

http://forums.eviews.com/viewtopic.php?f=7&t=465

[mtaimoor23]

Thanks for guidance but it does not solve my problem.Actually I do not have x variable in my equation of arima (1,1,4)

[EViews Gareth]

Then the Excel file examples in that post should be just what you need.

[mtaimoor23]

dear as I said before my data don not contains any x variable but the excel contains.Will you please explain the process without x variable.

[EViews Gareth]

Just take out the X variables from the example.

[mtaimoor23]

@EViews Gareth I apologize to say that still i could not get correct manual forecasts with out x variable,Kindly post the solution on any dummy data which reconcile both EVIEWS and Excel results.

[EViews Gareth]

Code: Select all

rndseed 1 create u 100 series x=nrnd series y=nrnd equation e1.ls(z) y c ar(1) ma(1) 'Static Forecast series e=0 smpl 2 100 e = y-c(1)-c(2)*(y(-1)-c(1)) - c(3)*e(-1) series yhats2=c(1) + c(2)*(y(-1)-c(1))+ c(3)*e(-1) 'completely manual calculation e1.fit yhats1 'EViews fitted values (static) show yhats1 yhats2