insurficient number of observations problem

For questions regarding the import, export and manipulation of data in EViews, including graphing and basic statistics.

Moderators: EViews Gareth, EViews Steve, EViews Moderator, EViews Jason

newtoeviews
Posts: 7
Joined: Fri Jan 23, 2015 8:51 am

insurficient number of observations problem

Postby newtoeviews » Fri Jan 23, 2015 9:12 am

Hi all,

First of all, I have been at the moment quite often here for my master thesis, which requires me to analyse data set. I have as well read topics here related to technical problems and really appreciated all the questions and answers. I actually get the same problems as some people here with "insurficient number of observations" problem. Apprencially, I have checked it again after I read topics here. All my data are correct, no missing observations.

My model is: log(y) = log(x2) log(x3) log(x4) x7 x8 x9

x7, x8, x9 as dummies.

I have panel data set from 2000 to 2013, 84 countries. So I have in total 84x14 = 1176

First, I run the command ls log(y) log(x2) log(x3) log(x4) (x7) (x8) (x9) and got the problem. Then, I managed to run it with only y= x2 x3 x4 x7 x8 x9, I got the result. Or even seperately only with log(y) log(x2); then log(y) log(x3) and so on, the result shows up. It therefore turns out that, I think that I get no problem with data but somehow more with technical things that I really have no clue.

Could you guys please help me to work it out? I really appreciate your help because I am sick of it for the whole week.

With regards, HM

startz
Non-normality and collinearity are NOT problems!
Posts: 3796
Joined: Wed Sep 17, 2008 2:25 pm

Re: insurficient number of observations problem

Postby startz » Fri Jan 23, 2015 9:18 am

Are some of your variables zero or negative?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13583
Joined: Tue Sep 16, 2008 5:38 pm

Re: insurficient number of observations problem

Postby EViews Gareth » Fri Jan 23, 2015 9:19 am

Without seeing the workfile, we can only guess. But if I were to guess, I'd say that y, x2, x3 or x4 have zero or negative values.

newtoeviews
Posts: 7
Joined: Fri Jan 23, 2015 8:51 am

Re: insurficient number of observations problem

Postby newtoeviews » Mon Jan 26, 2015 10:18 am

Thanks for your relies. I attach here by the way the workfile. I actually set all zero values to 1 so that it was supposed to work well, but it was not. Hope you can find the problem here.

I really appreciate your response.

Thanks, HM
Attachments
final.wf1
(139.75 KiB) Downloaded 437 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3796
Joined: Wed Sep 17, 2008 2:25 pm

Re: insurficient number of observations problem

Postby startz » Mon Jan 26, 2015 10:22 am

Works fine using the workfile you posted.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13583
Joined: Tue Sep 16, 2008 5:38 pm

Re: insurficient number of observations problem

Postby EViews Gareth » Mon Jan 26, 2015 10:32 am

Works fine using the workfile you posted.
What he said.

newtoeviews
Posts: 7
Joined: Fri Jan 23, 2015 8:51 am

Re: insurficient number of observations problem

Postby newtoeviews » Mon Jan 26, 2015 10:34 am

I have checked again with log(y) log(x2) log(x3) log(x4) and it gave result but for the whole equation, including x7 x8 x9, it won´t.

startz
Non-normality and collinearity are NOT problems!
Posts: 3796
Joined: Wed Sep 17, 2008 2:25 pm

Re: insurficient number of observations problem

Postby startz » Mon Jan 26, 2015 10:38 am

Is this what you did?
Capture.PNG
Capture.PNG (28.96 KiB) Viewed 7864 times

newtoeviews
Posts: 7
Joined: Fri Jan 23, 2015 8:51 am

Re: insurficient number of observations problem

Postby newtoeviews » Mon Jan 26, 2015 10:45 am

There is perhaps some problem of my eviews because I still can not get it. But thank you so much. I know now my data is all correct.

Many thanks and best regards, HM


Return to “Data Manipulation”

Who is online

Users browsing this forum: No registered users and 1 guest