Hi,
I am using the quadratic trend estimation technique to get trend GDP. My regression equation is:
GDP = β0 + β1Time + β2Time^2
when I run this regression the output from the estimation gives me the coefficient, std error, tstats and prob for each of the regressors.
Since this estimation or output will represent the trend GDP, I will need a time series of this estimation in order to measure the output gap.
Note the output gap is measured as: actual GDP minus trend GDP.
My question is therefore how do I get a time series from the quadratic trend estimation? I need a time series of the trend data which was estimated so I can subtract it from the actual GDP data to get the output gap.
I checked the eviews manual but I did not get a solution so any assistance will be appreciated. Thanks.
Using Quadratic Trend to estimate trend GDP
Moderators: EViews Gareth, EViews Moderator

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 12509
 Joined: Tue Sep 16, 2008 5:38 pm
Using Quadratic Trend to estimate trend GDP
I think you just need to use the fitted values from the equation, which you can do by hitting the forecast button.
Follow us on Twitter @IHSEViews

 Nonnormality and collinearity are NOT problems!
 Posts: 3567
 Joined: Wed Sep 17, 2008 2:25 pm
Re: Using Quadratic Trend to estimate trend GDP
What Gareth said.
Or inside the sample period
By the way, this is probably not a good way to model trend GDP.
Or inside the sample period
Code: Select all
series outputgap = resid
By the way, this is probably not a good way to model trend GDP.
Re: Using Quadratic Trend to estimate trend GDP
Thanks Gareth. Can I also get this fitted series by going to View/Actual, Fitted, Residuals from the workfile?
Re: Using Quadratic Trend to estimate trend GDP
So I know this was posted years ago but Ive just stumbled on it and one thought for anyone else reading this now is that there is the idea that GDP could be difference stationary rather than trend stationary
Who is online
Users browsing this forum: No registered users and 16 guests