Announcement

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

  • interpretation of "athrho" in a bivariate ordered probit

    Could some explain to me how to interpretate the "athrho" coefficient after a bivariate ordered probit regression?
    If you can make a response in two times: i. technical ii. non-technical.
    Thanks in advance.
    Regards,
    XR

  • #2
    See this Stata tip, which is now freely available: https://www.stata-journal.com/articl...article=st0230
    ---------------------------------
    Maarten L. Buis
    University of Konstanz
    Department of history and sociology
    box 40
    78457 Konstanz
    Germany
    http://www.maartenbuis.nl
    ---------------------------------

    Comment


    • #3
      Dear Sir, thank You for your return.
      I understand, reading your paper, that athrho is a transformed version of a parameter.
      But I do not manage to understand the signification of the parameter.
      Could you please take a look at my one-page joint result, so as to help me to interpretate?
      Thanks in advance.
      Regards,
      XR
      Attached Files

      Comment


      • #4
        Please do read the FAQ. There you'll find important information. For example, there is a recommendation to not to post attachments.

        Since you're a new member, I decided to share the information from the FAQ, specifically about this matter, but reading the whole FAQ is fully worth the time. .

        12.5 Posting attachments: please don't...

        There are several "please don't" requests here, but good reasons for them all.

        Please do not post .gph files, as they can't be read without flipping back and forth between Stata and the forum software, thus making your posts much more difficult to follow.

        In particular, please do not post screenshots. Many members will not be able to read them at all; they usually can't be read easily; and they do not allow copy and paste of data or code, which is highly desirable to allow experienced members to make precise suggestions for your questions.

        You are asked not to post attachments that are in Word or Excel file formats (.doc, .docx, .xls, .xlsx), because
        • many members just don't have or don't use such software
        • obliging other members to open those programs to see your problem is at best awkward and indirect
        • many members have zero-risk policies on not opening such files from third parties
        • it will often not be clear exactly how such data will appear in Stata, or even whether they can be read successfully into Stata
        • there are better ways to show the information, as explained just above.
        Finally, we ask that in general you don't post .dta or .zip files either. This is because
        • as above, it obliges members to fire up Stata (and/or some other program) with your file to see the problem, which could be difficult or time-consuming if you have a large or complicated dataset
        • members may have versions of Stata earlier than yours such that they can not read your .dta files anyway
        • threads become more difficult to understand if they depend on people reading in a dataset: short code and data examples are much easier to work with, as explained above.
        Best regards,

        Marcos

        Comment


        • #5
          athrho is the transformed version of rho, the correlation between the error terms. So if you are interested in that you just look at the rho reported rather than the athrho. Why athrho apears in the output is briefly explained in the article.
          ---------------------------------
          Maarten L. Buis
          University of Konstanz
          Department of history and sociology
          box 40
          78457 Konstanz
          Germany
          http://www.maartenbuis.nl
          ---------------------------------

          Comment


          • #6
            I am exploring biprobit to model correlation between related outcomes (physical abuse and emotional abuse).
            I read the paper by Maarten Buis which clarifies that rho relates to the correlation between regression residuals (https://www.stata-journal.com/articl...article=st0230)
            I wanted to clarify whether the rho in the biprobit output also relates to correlation between regression residuals?
            This paper by Fieuws and colleagues made me think that rho in biprobit output might refer to correlation between random coefficients. (https://rss.onlinelibrary.wiley.com/...6.2006.00546.x).

            Comment


            • #7
              The rho in the output is the correlation between the error terms.

              It is a technical issue that Stata (and most other programs) don't estimate rho directly but rather Fischers' z transformation of rho, only after the estimation is done is the estimate back-transformed to the correlation metric. Usually, you don't need to care about this, but there are special situations where this can be useful to know. However, if you just want to know what the estimated correlation between the error terms is, then just looking at rho will give you the answer.
              ---------------------------------
              Maarten L. Buis
              University of Konstanz
              Department of history and sociology
              box 40
              78457 Konstanz
              Germany
              http://www.maartenbuis.nl
              ---------------------------------

              Comment


              • #8
                Thank you for that clarification Maarten Buis . In my current analysis I am attempting to account for correlated outcomes. Would you be able to briefly explain, or point me to a good explanation, of how correlating the error terms accounts for correlated outcomes?

                Comment

                Working...
                X