Announcement

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

  • Bayesian Model Average

    Hi, I have recently ran xtlogit models and to further get better estimates I assume from my understanding is to use BMA on the models I have generated. My issues is how to specify the BMA on my models...can you please advise? Thanks

  • #2
    Shamsuddeen:
    see https://www.stata-journal.com/sjpdf....iclenum=st0239
    Kind regards,
    Carlo
    (StataNow 18.5)

    Comment


    • #3
      Thanks alot...ll go through it and hopefully resolve my issue. Really appreciate it.

      Comment


      • #4
        Originally posted by Carlo Lazzaro View Post
        Thanks again. I have read the paper...the application of BMA in it was for linear regression, however my issue is with the application of BMA to a panel data. I have looked at a number of resources but havent found any reference to applying BMA to panel data. Can you please help? Thanks

        Comment


        • #5
          Hi Carlo, apologies the above( the paper already answered this), what I meant to ask is my dependent variable is a crisis binary variable....does it require using a different something different or its sufficient to estimate it using the bma prefix and use the estimates as they are for my final analysis? Thanks

          Comment


          • #6
            Shamsuddeen:
            see: http://bms.zeugner.eu/resources/
            Kind regards,
            Carlo
            (StataNow 18.5)

            Comment


            • #7
              Thanks for pointing me to all the resources...really helpful. I have looked at quite a bit so far, while really informative and useful....the resources have not answered how I can use BMA panel logit...Perhaps am asking a lot but ll appreciate the help.

              Comment


              • #8
                Hello everyone,

                I am trying to apply Bayesian Model Averaging (BMA) for panel data using Stata 18. I first use the codes "xtset id year", then "bmaregress y x1 x2 .... x12, mprior(uniform) gprior(uip)".
                However, I have noticed that Stata does not recognize that my dataset is panel. I could not find any way to do that and would be grateful if you have any suggestions.
                Eric KOUAKOU

                Comment


                • #9
                  Eric:
                  welcome to this forum.
                  What does it mean that
                  Stata does not recognize that my dataset is panel
                  ?
                  Did Stata provide you with a message about this issue?
                  Kind regards,
                  Carlo
                  (StataNow 18.5)

                  Comment

                  Working...
                  X