outlier dummy

[Mathias]

Hi,

I have a problem concering the generation of an outlier - dummy in Eviews 7.
After I've made an panel least squares regression and I would like to correct the equation for extreme observations. This by creating an outlier - dummy

Therefore I've opened the equation window of the panel least squares regression and made residual series.
By doing: proc - make residual series --> name: resid_ols

Subsequently,
Generate series by equation:
- Outlier = 1 with the sample = 2001 2010 if ap>0 and emp>0 and tw>0 and mva>0 and leeftijd>0 and resid_ols<-3 or resid_ols>3

Then I saw that the observation 204 in the year 2001 is an extreme observation because it has a value of 1.

Now when I want to correct for this only extreme observation:
Generate series by equation:
- outlier204_01=0 with the sample = 2001 2010 if ap>0 and emp>0 and tw>0 and mva>0 and leeftijd>0
Then
- outlier204_01=1 with the sample= 2001 2010 if ap>0 and emp>0 and tw>0 and mva>0 and leeftijd>0 and id=204 and year=2001

When I check the output form this outlier dummy, I see that the observation 204 in the year 2001 doesn't get a value of 1.
I don't know what I did wrong, perhaps I have to delete the specific observation manually?

Thanks in advance!!

[EViews Gareth]

Hard to tell without seeing your data.