"Variable not defined"

[cyclechic]

Hi guys,

Thank you in advance for your help. I'm in the (very) early stages of my analysis and am completed defeated by this one issue.

I'm running a panel data regression of 19 countries (the first being labelled 'alg' for Algeria) where Corruption is my dependent variable (entered into EViews as cor_?). After inputting all my data (correctly as far as I am aware, I have checked several times) I am trying to run the simplest of OLS regressions, inputting 'cor_?' as the dependent variable and c, nat_?, gdp_?, opn_?, edu_? and law_? as my various independent variables. I click 'OK' and the message 'cor_alg is not defined' appears on screen, thus taking me back to the beginning and not producing the output I require.

I presume this is a fairly elementary issue, however I just cannot work out what this message means/where I'm going wrong/why I am presented with this. Any help really would be much appreciated.

Thanks so much.

NB EViews 7 is the version I'm using!

[EViews Gareth]

It seems like you're mixing terms.

Do you have a panel data workfile? If so, you don't need to use the ? notation.

If you don't have a panel workfile, and, rather, are using a Pool object to perform the analysis, then you'll need to use the Pool object itself to run the estimation.

See here: http://forums.eviews.com/viewtopic.php?f=7&t=3784

[cyclechic]

Thank you for your reply.

I'm not using the panel data workfile, but am using the pool object to perform the analysis, in so much as I 'created a New Object' and then entered the names of each of my cross-sections (_alg, _bol etc.). I am highlighting my pool object and choosing 'Estimate' from the task bar and entering my dependent and independent variables when I am confronted with the undefined variable message.

[EViews Gareth]

Could you provide the workfile?

[EViews Glenn]

I'm going to ask a stupid question...

Do you have a series named COR_ALG in the workfile?