Announcement

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

  • 3SLS with same independent variables for equations

    I want to estimate a system of 2 equations with endogeneous variables:

    reg3 (y1 x1 x2 x3 x4) (y2 x1 x2 x3 x4) , endog(x1 x2) exog(lag_x1 lag_x2 lag2_x1 lag2_x2)

    So the independent variables of both equations are identical: it is highly likely that the error terms of the two equations are affected by the same exogeneous shocks and therefore related.

    Does my specification make sense? My problem is that I dont see any difference in the results compared to a 2sls estimation....

    Many thanks


  • #2
    You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data.

    When you say results are identical, do you mean really identical or just very similar? [Providing "readable Stata output" would have helped here.] Both 2SLS and 3SLS are consistent although they usually give slightly different results.

    Comment

    Working...
    X