The forecasting process

[ESVAL2016]

Hi, for all
how to convert the differenced forecast series to the original series !!

Thanks for your help.
Regards,,

[EViews Gareth]

We'll need more detail.

[ESVAL2016]

Thanks a lot for your help.
When I started with my series, I found my series is non-stationary.
I converted my series to be stationary by using:: diff.
And, I found the fit model in this case and I did the forecasting process for the differenced series.
Now, I want to find the forecasting for the original series.

Thanks for your time.

[EViews Gareth]

By default if you estimate an equation where the dependent variable is a difference, it will forecast the undifference values.

[EViews Gareth]

http://www.eviews.com/Learning/forecasting.html

Slide 44 onwards.

[ESVAL2016]

Thanks a lot for your cooperation.
But here,
If the model includes a lagged dependent variable. It matters for forecasting purposes whether the dependent variable is an auto series or not.
My series not Auto series, and the differenced series here as ordinary series.

Regards,

[ESVAL2016]

So, in this case, we can write the first difference in the estimate equation as d(1) or not?? in order to work directly with the original series.
What is your opinion??

For example:


Thanks for your help.

[EViews Gareth]

I don't understand your question. Just write the dependent variable as d(myseries). You can include lags as independent regressors if you want - it doesn't change anything.