EViews.com

Good Morning,
I need to work with a panel data with three digits (country, industry , time). I know that fixed effects are just cross section dummies and that I can use the command @expand(country) @expand(industry, dropfirst). But how I have to structure the database ? should the workfile be unstructured ? What next?

A practical suggestion will be highly appreciated
Alfonso Arpaia

Depends on what you want to do with the data.

Thanks for your reply. I want to study the effect of R&D on productivity using data by industry, county and time.
Therefore I need to run a regression of the type

y(c,i,t)= a(c)+b(i)+g(t)+ B' TFP(i,c,t)+ u(i,c,t)

where a,b and g are fixed effect by country, sector and period
d is the coefficient capturing the effect of R&D on TFP.

I used intensively Eviews but never worked with three identifiers. Since the programme does not have a build in option I would need to estimate with LS with dummies . but I don't know how to set up the data base for this purpose
Thanks for your help

For running the simple regression, it doesn't really matter, so I'd just leave it unstructured.