Page 1 of 1

TSLS "not defined" notification

Posted: Wed Mar 07, 2012 10:49 pm
by lalalaguy
when i try to run TSLS equation, it says that the dependent variable is "not defined".
LN_GDP is the dependent variable, there are 3 explanatory variables and 2 instrumental variables
I input in "equation specification"
ln_gdp c lat_90 lcopen control_of_corruption

in "instrument list"
c lat_90 logfrankrom logem

I am new to eviews and my question may be very simple, but please tell me what to do

TSLS "not defined" notification

Posted: Thu Mar 08, 2012 12:02 am
by EViews Gareth
Is there a variable called ln_gdp in the workfile?

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 11:26 am
by lalalaguy
thank you for your reply!
Yes, there is a variable ln_gdp. It's natural logorythm for gdp per capita for a number of coutries.
Is there something wrong with the name itself?
I'm using eviews 7, by the way

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 11:32 am
by EViews Gareth
Post your workfile.

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 11:56 am
by lalalaguy
thank you for your help, here is the workfile.
i enter the parameters in the equation window as i wrote in my initial post.

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 12:11 pm
by EViews Gareth
The LN_GDP variable has a funny character (it isn't a true _ in the name, it is some weird other character). Rename that series to something else.

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 12:21 pm
by lalalaguy
I've just renamed it and other variables to eliminate all "_" signs. But it is giving the same error message. I'm starting to think there is something wrong with the data series themselves. I'm reuploading the workfile with the new names. It looks like this now:


C
CORRUPTION
COUNTRY
LAT90
LCOPEN
LNGPD
LOGEM
LOGFRANKROM
RESID

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 12:34 pm
by EViews Gareth
LNGPD is not the same as LNGDP :D

Re: TSLS "not defined" notification

Posted: Thu Mar 08, 2012 12:41 pm
by lalalaguy
LNGPD is not the same as LNGPD :D
oh, thank you very much!
i can't believe i made such a silly mistake and it cost me so much time.

Re: TSLS "not defined" notification

Posted: Wed Mar 14, 2012 12:47 pm
by nicoles
Hi,

I am having the same problem. I've checked to make sure _ is not __ and looked at the spelling, but just cant work it out. I am trying to estimate equation lgdp? = c ls? ln? lpr? and eviews says that lgdp is not defined. Can you help?

Nicole

Re: TSLS "not defined" notification

Posted: Wed Mar 14, 2012 1:30 pm
by startz
That's probably because in the file you posted lgdp is not defined.

Re: TSLS "not defined" notification

Posted: Tue Apr 02, 2013 6:44 pm
by mcvix
Hi!
I am new to eviews and I am happy I found this forum.

I receive the same `not defined` notification. Both my independent and dependent variables are matrices.

Is it perhaps the reason. If yes, how can I run a co-integration regression on eviews then. In the files below,

fbeta is my dependant variable and rbeta is my independent variable.

I would deeply appreciate your help.
fbeta.wf1
(38.72 KiB) Downloaded 187 times
rbeta.wf1
(38.72 KiB) Downloaded 181 times