Re: PMG questions
by Lars Ahnland
Great, thanks. Yes, ofc lags of both dep and indep variables should be used, I see that now. What about the contemporary, non-lagged, version of the independent variables, are they included?
/Lars
Skickat från Outlook för Android<https://aka.ms/AAb9ysg>
2 years, 4 months
- 1
- 0
Loop Construction Command
by Dhanasekaran Kuppusamy
Dear Gretl Experts,
I worked with a sample consisting of 6
regions (1,2,……,6) with 2 areas (1 &2) and thus, a total of 12
sub-samples. But only 11 subsamples are available for estimation. I used
the following loop for the estimation. For all the 12 sub-samples (if
available) the following loop works well. When one sub- sample is
unavailable, I used the following code and got the message “No
observations would be left!”.
Code:
discrete Region
discrete Area
loop i=1..6 # Six Regions
loop j=1..2 # Two areas
smpl (Region==i)&&(Area==j) --restrict --replace
Message:
No observations would be left!
smpl (Region==i)&&(Area==j) --restrict --replace
loop j=1..2
In addition to the above, I want to create a combined loop in order to
compute the Region-wise combined area (1+2) and for a full sample( six
more sub-samples plus 1 more full sample in the same loop and thus a
total of 19 subsamples).
I would be thankful to you, if you could give a solution to the above issue
and to compute the Region-wise combined areas (1+2) and for a full sample
in addition to the loop possibly.
Thanking You,
Dr.K.Dhanasekaran
2 years, 5 months
- 2
- 3