From panel to pooled

[krest2000]

Dear all,

I have got a panel data set, containing 55 developing countries and 15 variables. However, I need to perform a pooled regression. Thus, I suppose I need to transform the data in some manner?

I created the data set in Excel, and then opened it in E-views as a Dated Panel - should I do that or is there some other way? Then I clicked: view>new..>pool and wrote the name of all cross sections in a list starting each name by _.

But, the remaining issue is that the series are still "panel-series", e.g. I only have one series for inflation, not inflation_germany etc.

What should I do? I am quite new to this, and to be honest, I really do not understand what step I should take next.

In summery: I have panel data, but I want to do a pooled regression. What should I do?


If anyone can help me with this, I am truly grateful!!

[EViews Gareth]

From a panel workfile you can convert to a pool by clicking on Proc->Reshape Current Page->Unstack in New Page.

Enter your cross-section ID series in the first box ("Series containing unstacking identifiers"). Enter "@date" in the second box. Hit ok. Done.

[krest2000]

Thanks!

Sounds easy enough. However, I seem to be doing something wrong anyway.

I followed your instructions. But got the message "series specified for unstacking does not have valid values for forming series names".

What am I doing wrong?

[EViews Gareth]

When you go from panel to pool you get series created that contain the name of each cross-section. Thus your cross-sections must contain only valid EViews name characters. No spaces, no punctuation.

[krest2000]

Thanks!!