Announcement

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

  • Thresholds

    Hi everyone

    I want to investigate from wich debt rate the labor productivity will start decreasing to determine the optimal debt level. I know that I need to implement debt tresholds but wich commands do I therefore have to insert?

    I have already had the following advice:

    "If it is your expectation that labor productivity sometimes increases with debt rate and sometimes decreases with debt rate depending on the value, then you need a different, non-linear, model."

    It is important that the commands will be available in Stata version 12.1 because I am not in the possibility to use another version of Stata.


    Thanks in advance.

  • #2
    Stata 15 has threshold regression. It far from my field, but I gather this is what you wish. Unfortunately, I don't know if there is a user-written program which can be used in earlier version.
    Best regards,

    Marcos

    Comment


    • #3
      Joachim asked this earlier on https://www.statalist.org/forums/for...ebt-thresholds I advised him to ask again in case people didn't engage with "Debt".

      Comment


      • #4
        I am now looking to buy Stata version 15. But there are different types:
        - Stata/IC
        - Stata/SE
        - Stata/MP

        Can I run threshold regression on all these different types? The plan is to buy type "Stata/IC".

        Comment


        • #5
          The flavours are basically related to the amount of variables as well as the speed. You may found a thorough description in the StataCorp’s webpage.
          Best regards,

          Marcos

          Comment


          • #6
            Thank you for the answer.

            But is there nobody who knows how you can run thresholds in Stata version 12.1?

            Comment


            • #7
              It would seem that threshold regression (-threshold- command) was introduced in Stata 15. I could not find reference to it in the Stata 14 manual.

              Comment


              • #8
                I also find no information about it, but my supervisor of school says that it is possible to run threshold regression in Stata version 12.1.

                Comment


                • #9
                  I have bought Stata version 15.1. I want to run the next threshold regression:


                  threshold logarbeidsproductiviteit, threshvar(logschuldgraad) regionvars(logomzet lognettotoegevoegdewaarde logschulden logarbeidsintensiteit logloonkost logaantalwerknemers)

                  I get the error: (r198).

                  What do I need to change?

                  Comment


                  • #10
                    Dear Joachim, If you are using cross-sectional data, please consider https://www.ssc.wisc.edu/~bhansen/pr...threshold.html, or if you are using panel data. please check this out https://www.stata-journal.com/articl...article=st0373.
                    Ho-Chuan (River) Huang
                    Stata 17.0, MP(4)

                    Comment


                    • #11
                      I am already using that article, but the problem is the error (r198).

                      Comment


                      • #12
                        Dear Joachim, Are you replying to my suggestions? If so, then we are talking about different things as far as I can see. I don't use the "default" threshold command.

                        Ho-Chuan (River) Huang
                        Stata 17.0, MP(4)

                        Comment


                        • #13
                          I want to use that article for the description of the analysis. But Indeed, i am using the default threshold command.
                          Last edited by Joachim Vanderheyden; 18 Mar 2019, 04:06.

                          Comment


                          • #14
                            I'm just wondering whether you could inform it the data was - tsset - or not.
                            Best regards,

                            Marcos

                            Comment

                            Working...
                            X