Announcement

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

  • Can someone who has the old Stata 7 message me and run a very short do file?

    I am trying to replicate data in a paper, and the results I got was different from the data in the paper in a special way that they were all powers of 2, e.g. 64, 2048, 8.
    I suspect this happens due to the fact that data in the paper was generated using Stata 7.
    Can someone who has Stata 7 message me and run the do-file for me to make sure?
    Thanks!

  • #2
    Stata 7 is pretty old. It was released in 2000. It may even be old enough that it won't run on contemporary hardware, I don't know. If there are any copies still floating around, this Forum is probably the best place to look for them.

    That said, have you tried running the do-file on your current Stata using version control?
    Last edited by Clyde Schechter; 05 Oct 2022, 22:53.

    Comment


    • #3
      I didn’t know version control was a thing. I’ll give it a try. Thank you!

      Comment


      • #4
        FWIW, I used Stata 7 when it was a thing. If anything produced only powers of 2 in Stata 7, I would lay overwhelming odds on it resulting from the user's code, not from a limitation or bug in Stata 7. So, I don't think it's Stata 7 you need. It's to show us the do-file.

        Comment


        • #5
          Actually its not that my do file only produced powers of 2.
          Let's say that the author of a paper made original.do and originaloutput.dta available on the internet and I downloaded them.
          "originaloutput.dta" contains the output variable "output_variable_original".
          And the author said original.do produces originaloutput.dta.
          But once I run original.do, and let's say I chose the name myoutput.dta,
          which contains the output variable "output_variable_my",
          the differences between "output_variable_original" and "output_variable_my" are all powers of 2 like 0.25, 0.5, 1, 2, 4, 8, 16, etc.

          Comment


          • #6
            Put yourself in our position. All we know is that there is some unspecified code and some unspecified dataset and the result of the code does not match the dataset. I would love to have magical abilities that would enable me to find the problem with just that information, but unfortunately I was only trained at a university and not at Hogwarts. In short, without the code and the data, there is nothing we can do.
            ---------------------------------
            Maarten L. Buis
            University of Konstanz
            Department of history and sociology
            box 40
            78457 Konstanz
            Germany
            http://www.maartenbuis.nl
            ---------------------------------

            Comment

            Working...
            X