Announcement

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

  • Bootstrapping for Panel Data Quantile Regression

    Dear Statalists,

    I'd appreciate if you could tell me which command I should use to run a quantile regression for panel data FE using bootstrap.
    So far, I only know hot to run the model using the qregpd and I'm applying the following command:

    qregpd CRS X1 X2 X3 X4 X5 X6, id(id) fix(year) optimize(mcmc) noisy draws(1000) burn(100) arate(.5) quantile (25)


    Thanks in advance!

  • #2
    Hi Renata
    First of all, Not all commands do the same kind of estimation.
    qregpd, for example, doesn't really control for fixed effects as we are used to (adding dummies for example), but rather controls for them as if controlling for the distribution of the fixed effects.

    Now, Machado and Santos Silva (2019) https://www.sciencedirect.com/scienc...04407619300648
    does suggest an option for Qregressions with panel data that is easy to understand.

    He implements this with -xtqreg- available from SSC

    I also have the command -mmqreg- (also from ssc) that allows for multiple fixed effects.

    You should be able to use the prefix bootstrap in both commands
    HTH

    Fernando

    Comment


    • #3
      Dear Renata Benicio,

      Adding to Fernando's helpful advice, you can see an example of how to use bootstrap in this context (and how to implement a jackknife correction) here.

      Best wishes,

      Joao

      Comment


      • #4
        Thanks FernandoRios thanks for the advice! I'll check it out!

        Joao Santos Silva Thanks! that example is very useful. I'll try to adapt to my data

        Comment


        • #5
          Hello @Joao Santos Silva , @FernandoRios
          I'm trying to do the same thing but I don't know much about code, a do-file would be more useful

          Comment


          • #6
            In #3 I provide a link to a do file.

            Comment


            • #7
              In my reading I understood that the Bootstrap is more sophisticated than the jackknife it would be my pleasure to have one with the Bootstrap method itself
              Best regards

              Comment


              • #8
                Check the line stating with
                Code:
                 
                 qui bs

                Comment


                • #9
                  to apply this for my data set do i just have to change variables name?

                  Comment


                  • #10
                    please Sir what does the code actually do

                    Comment


                    • #11
                      Please see the help file for the bs (bootstrap) command.

                      Comment


                      • #12
                        what the code really does

                        Comment


                        • #13
                          Originally posted by kwenga loic View Post
                          to apply this for my data set do i just have to change variables name?
                          what about this ?

                          Comment


                          • #14
                            Click image for larger version

Name:	Capture.PNG
Views:	1
Size:	41.4 KB
ID:	1708017


                            Good Morning sir i run That command and i got that i dont why ! thanks for help

                            Comment


                            • #15
                              Please send me your data and code by email and I'll check it.

                              Comment

                              Working...
                              X