Fixed Effects, dummy dependable important


New Member

I am trying to analyse the effect of the variable VC (see image, dummy) on the variable HC and I've set up the database with other variables which will serve as more of control dummies and side analyses. The database is consistent of unbalanced panel data, however I've added blanks to the data so it becomes "balanced" and can be imported into my stat program (Gretl). While trying the fixed effects model I've disregarded that the dummy will be omitted due to obvious exact collinearity. Since the fixed effects estimator takes out all the variance at the group level, there is nothing left for those dummies to explain. But I am interested in knowing that exact effect. Is this solvable (maybe through transformation of my dataset)? If so, how?

A screenshot of the database composition is added below.

Kind regards

Side information:

VC -> Venture Capital, dummy variable which indicates whether or not the cross sectional variable has been a subject to Venture Capital injection;
HC -> HeadCount, number of employees during a given year;
IY -> Investment Year;
PE -> Payroll Expenses;
Bubble -> Dummy 1 if the VC investment was done before the bubble;
Crisis -> Dummy 1 if year = 2008;
VA -> Value Added;
NP -> Net Profit;
IPO -> Dummy 1 if company has ever issued an IPO.

Transformed database from VICO dataset ( As you can see, the data does not line up perfectly.

If more information is required, please do ask.