Announcement

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

  • midas: Hessian has become unstable or asymmetric r(504);

    I have one more query on midas please.
    While doing analysis, an error statement came up as below
    Hessian has become unstable or asymmetric
    r(504);

    When I clicked on r(504), the following message came up.

    Search of official help files, FAQs, Examples, and Stata Journals

    [P] error . . . . . . . . . . . . . . . . . . . . . . . . Return code 504
    matrix has missing values
    This return code is now infrequently used because, beginning
    with version 8, Stata now permits missing values in matrices.

    (end of search)
    In my database, there are no missing values.
    Please help

  • #2
    Dear Shripada,
    I was wondering whether you have any update regarding this error?
    Thanks and regards

    Comment


    • #3
      Dear Nicoletta, Apologies for the delayed reply because I had not logged in for many weeks. I found that if you add or subtract a value of one of the variables by one (example write 16 instead of 15), usually it resolves. Not sure if it is the right thing to do though. But as long as it is described in the methodology section transparently, hopefully it should be ok. I am keen to know your experience and that of others on this issue and approach. Regards

      Comment


      • #4
        I have this data on which I want to obtain forest plot using midas
        author studyid tp fp fn tn
        BASKIN 1993 44 5 3 3 10
        U Warawdekar 2006 66 83 10 44 40
        Ssu-Han Wang 2019 302 19 0 1 10

        I have used this code which worked well while obtaing the plots
        midas tp fp fn tn, id(author) ms(0.75) ford fors bfor(dss)

        I have been getting the following error which I am not able to rectify.

        number of quadrature points is greater than the number of observations
        r(459);

        Please, help me out understanding it.

        Comment


        • #5
          Hi Akash, The most likely reason is that there are only three studies in your meta-analysis. Try adding a fourth study (dummy) data to see if it works. Best wishes.

          Comment

          Working...
          X