create industry dummy to control for industry in panel data

[Jodie Lewicki]

I need to control my data not only for board size, firm size and ROA and country of origin where I already created a dummy variable where Germany has the value of 1 and the US of 0, but I also need to control for industry type. I have however around 34 different industries. Can I still create a dummy variable for those and include them in my panel data regression? And if yes: How do I create a dummy variable for a lot of industries?

[Carlo Lazzaro]

Jodie:
1) you can go:

tab industry, g(new_ind)

2) a word of (unsolicited) advice: if you go -xtreg,fe- there's no gain in plugging -i.industry- in the right-hand side of your regression equation, unless (some) firms within the same panel are expected to switch from one industry to another as time goes by.

[Jodie Lewicki]

Thank you!

I used xtreg with -re- so far, so with the random effects model it should work right?

[Jodie Lewicki]

Moreover, I already have the value of 0 and 1 for the countries. Can I still use numbers 1 etc for this dummy variable then or does Stata automatically create different dummy variables for me?

[Carlo Lazzaro]

Jodie:
1) -i.industry- as a predictor is ok with -xtreg, re-;
2) you can code -i.industry- from 0 (or whatever) onwards: please note:
a) -fvvarlist- notation is by far the most effective and efficient way to deal with categorical predictors and/or interactions;
b) no matter the starting value of your code, Stata will recode internally from 0 onwards.