Hey
I'm working with pooled data with 29 cross-sections for the period Jan 2006 until Dec 2010 and with 18 explanatory variables. Now I've run a simple regression, but I noticed that a lot of the observations were not included. From 20,058 to 1,248 observations; see output
Dependent Variable: BASPRD2?
Method: Pooled Least Squares
Date: 09/02/11 Time: 15:06
Sample (adjusted): 1/31/2006 12/30/2010
Included observations: 1248 after adjustments
Cross-sections included: 27
Total pool (unbalanced) observations: 20058
Cross sections without valid observations dropped
I was just wondering how EViews decides whether an observation is included or not. Is it the case that if one datapoint for 1 variable at a certain day in 1 of the cross sections is missing, this particular day in all the 29 cross-sections will be dropped? So for example, if on January 1st 2006, data for the variable X is missing in the 3rd cross-section, then all the data for January 1st 2006 will be dropped in every cross-section, even though the remaining 17 variables do have datapoints on this day in all cross sections.
If this is the case, is it possible to change the manner that EViews uses to included data? Because now I've lost a lot of observations. I was hoping that EViews would only drop the 3rd cross section for Jan 1 2006, instead of all 29 cross sections.
I hope I've made my question is clear and someone could help me.
How does EViews determine whether an observation is included
Moderators: EViews Gareth, EViews Steve, EViews Moderator, EViews Jason
-
EViews Glenn
- EViews Developer
- Posts: 2682
- Joined: Wed Oct 15, 2008 9:17 am
Re: How does EViews determine whether an observation is incl
EViews will drop only the relevant observation unless you check the Balance sample option in your estimation dialog. So unless you have done so, we have used the unbalanced sample with the maximum number of observations possible.
Re: How does EViews determine whether an observation is incl
Hello,
I was wondering if there is a why that EViews can display descriptive statistics only for the subsample of observations included in a regression (after dropping those where certain values are missing for some variables etc).
Thanks
I was wondering if there is a why that EViews can display descriptive statistics only for the subsample of observations included in a regression (after dropping those where certain values are missing for some variables etc).
Thanks
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3797
- Joined: Wed Sep 17, 2008 2:25 pm
Re: How does EViews determine whether an observation is incl
After the regression, give the commands
Code: Select all
series includedObs = resid<>na
smpl if includedObs-
EViews Gareth
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13603
- Joined: Tue Sep 16, 2008 5:38 pm
Re: How does EViews determine whether an observation is incl
Smpl if resid<>na
Who is online
Users browsing this forum: No registered users and 2 guests
