Control Variables

[Luca Trivellone]

Hi guys,

Sorry for the "dummy" question (lol). I have to make a very simple model where i have to regress Transaction multiples (EV/EBITDA) on a set of dummy variables which describe the key figures of the deal (method of payment, Acquisition, minority stake ecc.). The data are obviously cross sectional. I have to make this regression with target sector variable as a control variable and size variable and I used a very simple sort function on Stata (that create a group of subsets based on the value of the sector variable). Do you think that is it wrong this procedure? Is there another procedure which is better than this one? Thanks for everyone who can answer

[Carlo Lazzaro]

Luca:
it is difficult to reply positively without seeing what you typed and what Stata gave you back (as per FAQ). You can share all that stuff via CODE delimiters (as per FAQ again).
That said, with all predictors being categorical variables (a more elegant way to name dummies) you may have perfect collinearity issues.
Eventually, dummy questions are rare, whereas questions searching for a helpful replies are more frequent.