Announcement

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

  • Merlin: error 3001 when attemtping to run model with family(ologit)

    Dear all,

    After several attempts to fix a recurring error when running the following commands within merlin, I am hoping to get som further guidance from the forum.

    For demosntration purposes, I'm using a public dataset: https://stats.idre.ucla.edu/stat/data/ologit.dta
    The problem appears to be independent of the actual dataset.

    Code:
    use https://stats.idre.ucla.edu/stat/data/ologit.dta, clear
    merlin(apply gpa, family(ologit))
    
    Fitting full model:
    
        merlin_logl_ologit():  3001  expected 1 arguments but received 3
            merlin_logl_ob():     -  function returned error
                 merlin_gf():     -  function returned error
          mopt__calluser_v():     -  function returned error
          opt__v0_calluser():     -  function returned error
        deriv__call1user_v():     -  function returned error
     _deriv__compute_value():     -  function returned error
                    _deriv():     -  function returned error
           opt__eval_nr_v0():     -  function returned error
                 opt__eval():     -  function returned error
    opt__looputil_iter0_common():     -  function returned error
    opt__looputil_iter0_nr():     -  function returned error
              opt__loop_nr():     -  function returned error
                 opt__loop():     -  function returned error
                _moptimize():     -  function returned error
               Mopt_maxmin():     -  function returned error
                     <istmt>:     -  function returned error
    r(3001);
    
    end of do-file
    -----------------
    The "traditional" ordinal logic regression runs w/o any issues:

    Code:
    .
    use https://stats.idre.ucla.edu/stat/data/ologit.dta, clear
    ologit(apply gpa)
    
    Iteration 0:  Log likelihood = -370.60264  
    Iteration 1:  Log likelihood = -366.30573  
    Iteration 2:  Log likelihood = -366.29998  
    Iteration 3:  Log likelihood = -366.29998  
    
    Ordered logistic regression                             Number of obs =    400
                                                            LR chi2(1)    =   8.61
                                                            Prob > chi2   = 0.0034
    Log likelihood = -366.29998                             Pseudo R2     = 0.0116
    
    ------------------------------------------------------------------------------
           apply | Coefficient  Std. err.      z    P>|z|     [95% conf. interval]
    -------------+----------------------------------------------------------------
             gpa |   .7248719   .2493023     2.91   0.004     .2362484    1.213495
    -------------+----------------------------------------------------------------
           /cut1 |   2.374854   .7569947                       .891172    3.858537
           /cut2 |   4.399911   .7814832                      2.868232     5.93159
    ------------------------------------------------------------------------------
    
    .
    end of do-file
    Thank you all for your help!



  • #2
    Michael Crowther

    Comment


    • #3
      This was a bug - now fixed in v2.4.5:

      Code:
      net install merlin, from("https://raw.githubusercontent.com/RedDoorAnalytics/merlin/main/")

      Comment


      • #4

        Michael Crowther

        Thank you so much for the quick fix.
        Works like a charm, now.

        Comment


        • #5
          also, see Statalist FAQ 3. What should I do before I post? "Before posting, search the forum for similar questions".

          Comment


          • #6
            Unless I'm missing something, Bjarte, this is a newly reported issue and newly fixed bug.

            Comment


            • #7
              Yes, this was a very fast fix of a new issue. I am sorry for my misplaced comment above.

              Comment

              Working...
              X