固定效应回归函数

Question

I read several questions about the fixed effects regression with felm() (package lfe).

I am wondering what the "rnorm(nlevels(..." does. My regression works when I write the following code:

spieler.eff=rnorm(nlevels(spieler))
sfeT1=felm(sätzegesp~TTRverf,spieler.eff|0|0, data=datregT)

But why do I need to call spieler.eff=rnorm(nlevels(spieler)) ?

I wrote the code like this because I've seen many others to do so. Actually I would have written

sfeT1=felm(sätzegesp~TTRverf,spieler|0|0, data=datregT) 

but this does not work.

So my question is, what the spieler.eff=rnorm(nlevels(spieler)) does.

Can anyone help me?

Thanks a lot in advance!

Answer 1

According to felm, the correct syntax for the formula is y ~ x1 | x:f1 | f2 | f3. You seem to have an extra comma after TTRVerf