thanks gareth
i have done first differencing and seasonal first differencing on my data. i have selected my best model for my first difference data and also selected a model for my seasonal first differenced data. i have used the models to forecast for a 4 month outside my range.
my question is how do i covert the forecast values from my models to comparable figures like my original data. because they only forecast differences for the 4 months because i differenced the data.
creating weekly and semi annual dummies with @expand
Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt
Re: creating weekly and semi annual dummies with @expand
thanks gareth
i have done first differencing and seasonal first differencing on my data. i have selected my best model for my first difference data and also selected a model for my seasonal first differenced data. i have used the models to forecast for a 4 month period outside my range.
my question is how do i covert the forecast values from my models to comparable figures like my original data because the models only forecasted differences for the 4 month period since my data was differenced.
i have done first differencing and seasonal first differencing on my data. i have selected my best model for my first difference data and also selected a model for my seasonal first differenced data. i have used the models to forecast for a 4 month period outside my range.
my question is how do i covert the forecast values from my models to comparable figures like my original data because the models only forecasted differences for the 4 month period since my data was differenced.
-
EViews Glenn
- EViews Developer
- Posts: 2682
- Joined: Wed Oct 15, 2008 9:17 am
Re: creating weekly and semi annual dummies with @expand
If you estimate your model with the differenced expression instead of a series generated from the differenced expression, the forecast procedure will give you the option of forecasting the original level. The docs have details.
Who is online
Users browsing this forum: No registered users and 1 guest
