Announcement

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

  • How to use instrumental variables estimation?

    Dear all,

    For my thesis, I need to run the following regression:
    Click image for larger version

Name:	regression1.png
Views:	3
Size:	17.5 KB
ID:	1548616



    In a paper, where I base my thesis on, it states: Instrumental variables estimation is still required because perf(it) is used to calculate both delta(perf(it+1)) and delta(perf(it)).
    For the moment, my regression in stata is in the following form (see attachment):
    Click image for larger version

Name:	Regression.png
Views:	1
Size:	15.5 KB
ID:	1548615




    The performance measure is OperatingIncome. The rest of the variables are slightly different from the specification above.
    My question is now: What model should I use to deal with this? I have read that the command ivreg2 can deal with this? Can someone propose an instrumental variable?

    Thanks in advance
    Last edited by Kevin Traen; 23 Apr 2020, 02:07.

  • #2
    Hi Kevin, you will be able to find an excellent answer to your question in David Roodman's Stata Journal article "How to do xtabond2: An introduction to difference and system GMM in Stata": https://journals.sagepub.com/doi/pdf...867X0900900106.

    Comment


    • #3
      To add to Hong's useful suggestion, there is a wide variety of appropriate estimators in Stata. You might consider ivreg, ivreg2, xtivreg, xtivreg2, xtabond, and xtdpdml. ivreg2, xtivreg2, and xtdpdml are user-written. From your equation, it looks like you have panel data, so you should look at the panel data estimators.

      You need to read up on selection and appropriate criteria for instrumental variables. In general, the choice of an instrumental variable depends on detailed understanding of your context. Depending on how long the serial correlation is, previous values of the variables may be used as instruments.

      Comment


      • #4

        Hello I have this code for system GMM
        Code:
        xtabond2 trisk l.trisk bind fenex femex feric crcomind femric fexric cbmeet l.rsanc inrel totd flev,  gmm (trisk fbric, lag(1  1) collapse eq(diff)) gmm(trisk fbric, lag(0 0) collapse eq(level)) iv(bind fenex femex feric crcomind femric fexri cbmeet inrel l.rsanc totd flev, equation(level)) robust orthogonal small nodiffsargan
        Can anyone help on how to implement same or similar modeling using xtivreg2?

        Comment

        Working...
        X