Friday, September 14, 2012

Dummy Variables - Again!

In a previous post (here) I had a few things to say about the dummy variables that we often use in regression analysis. I'm currently making changes to a related paper of mine that's at the "revise and re-submit" stage with a journal. So, to get further feedback, I presented the material in my department's Brown Bag seminar series earlier this week.

If you're interested, you can download the slides for that presentation from here.


© 2012, David E. Giles

7 comments:

  1. Just some basic questions:
    1) Is the use of dummy variables common in TSLS?

    2) When estimating a 4-equation TSLS, does estimating each equation and pasting to the model object of E-Views the same with "system" estimation command in E-Views?

    Thanks!

    ReplyDelete
    Replies
    1. 1. Yes.
      2. Yes, as long as you are declaring the same list of instruments when you estimate each equaiton separately.

      Good luck!

      Delete
  2. Thanks much for the information!

    ReplyDelete
  3. Wish I had seen that before submitting my dissertation, but I am happy to know that anyway. One question: What happens if we have a variable which is the interaction between one continuous and one dummy variable? Many thanks.

    ReplyDelete
    Replies
    1. The same problem will arise if the dummy takes just one (or a fixed, can't increase) non-zero value. The estimator for the coeficient will again be inconsistent.

      Delete
  4. Hi Dave,
    Awesome blog!
    Dave, I am looking some information about Monte Carlo simulation to generate a Probit model via MLE. For example I have y=(0,0,1,0,0)' and I would like to simulate 25 samples after 5, I mean y6,y7. Is it possible that you can give me some advice or useful information?

    ReplyDelete
  5. Thanks for the positive comment! Yes, I have some code from quite some time back. I'll dig it out and post it.

    ReplyDelete