Panel data: fixed effects model with many dummy variables

[student22]

Hi all,

Im writing my thesis about the performance of listed versus unlisted firms and have a large sample of panel data. I was thinking about using a Fixed Effects Model but have some questions about the model specification. My data consists of about 1000 firms and I created dummies for: status (listed or unlisted), country, and industry (8 sectors).

What I know about a fixed effects model and dummy variables:
- when you use no dummy variables: go to estimate equation> panel data> cross-section= fixed, so that it allows for heterogeneity among the firms
- when you use dummy variables: go to estimate equation> panel data> cross-section = random (and put your dummies in the estimation equation)
- when you use dummy variables, you use an intercept term and leave 1 category out (so in the case of the industry dummy, you have an intercept term and 7 industry dummies) or you use no intercept and use all 8 industry dummies

So, I have a lot of dummies, of which the status dummy (listed vs unlisted) is the most important one. Therefore I was thinking of the following model:
PERFORMANCE= C+ VARIABLE1+ DUMMYListed*VARIABLE1+ VARIABLE2+ DUMMYListed* VARIABLE2+ ...+ DUMMYcountry1+ DUMMYcountry2+..+DUMMYindustry1+...+DUMMYindustry8

QUESTION:
Im wondering whether this is the right specification as I use all dummy variables and have an intercept included as well. However, I need to know the effects of each country and each industry sector on the performance individually. So i'm afraid that when I remove one country and 1 industry dummy, the effects of this country and industry will be mixed within the intercept term.

Can someone help me with this problem? Thanks a lot!!!

[student22]

Just to be clear: I use interaction dummies for listed dummy variable with the independent variables of interest.