Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Panel regression with multiple endogenous variables

    Hello,

    I am struggeling a lot to find the right command to estimate the following regression for panel data:

    Correl = b1*T + b2*S + b3* F +b4*E

    where T, S and F are all endogenous variables and described by their own, differing instrumental variable terms.

    I figured out that in case I had a pooled OLS, I needed to use the command

    reg3 (Correl T S F E) (T I1 I2 I3) (S I5 I6) (F I7 I8), 2sls

    but unfortunately I cannot find the right command for a regression with multiple endogenous variables anywhere.

    Could someone please help me with that? I am really lost and this regression constitutes the basis of my dissertation which I will have to hand in soon....

    Thanks a lot!
Working...
X