Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Stata question

  • 16-11-2009 11:12pm
    #1
    Flamed Diving
    Closed Accounts Posts: 6,609 ✭✭✭


    Hello, I vaguely remember from my undergrad that there is a command that makes Stata X run as Stata X-1, and therefore could save files in this older version.

    In short, I have Stata 10 and want to save a *.dta as compatible for Stata 8.

    Can I do this?


Comments

  • #2
    [Deleted User]
    Posts: 5,589 ✭✭✭ [Deleted User]


    StatTransfer does that very handily, UCD should have it.

    CSV will also make the file transferable.


  • #3
    Flamed Diving
    Closed Accounts Posts: 6,609 ✭✭✭Flamed Diving


    Ok, a most bizarre thing is happening. But variables keep disappearing as I try to work on Stata 10 just now.

    Any idea why this might be? All I am doing is running simple panel models.


  • #4
    Économiste Monétaire
    Closed Accounts Posts: 2,208 ✭✭✭Économiste Monétaire


    Are you running an "official" version of Stata? If not, that's the problem ;).


  • #5
    Flamed Diving
    Closed Accounts Posts: 6,609 ✭✭✭Flamed Diving


    Yes, I am running an ;)fficial version of Stata 1;). I feel like writing a strongly worded letter.


  • #6
    [Deleted User]
    Posts: 5,589 ✭✭✭ [Deleted User]


    If you are using an incorrectly licenced version of stata or stat transfer, then that is a feature, not a flaw.

    Should you be running a perfectly legitimate version of stata, but you are unable to find your licence number, then try running it under linux. It isn't so forgetful in that environment. Ubuntu or Debian are Windows friendly linux distros.


  • Advertisement
  • #7
    Flamed Diving
    Closed Accounts Posts: 6,609 ✭✭✭Flamed Diving


    Ok, more questions. I am trying to pull off a fairly simple command here. This is version 9.1.
    gen logfdi = ln(usfdi)
    type mismatch
    r(109);

    The variable in the brackets is correct. I have run this type of command a trillion times before, so I can't see the problem.


  • #8
    [Deleted User]
    Posts: 5,589 ✭✭✭ [Deleted User]


    is it not log(varname)?


  • #9
    Flamed Diving
    Closed Accounts Posts: 6,609 ✭✭✭Flamed Diving


    Deleted User wrote: »
    is it not log(varname)?

    No, I even used Statas auto-generate feature to create the command and it still returned that error. However, I used a simple dataset that I have and tried the same command and it worked, so it seems specific to my dataset. The variable...

    :eek:

    *click, click, tap, tap*

    There is a letter in one of the cells. Always inspect the data, kids.

    EDIT: Nope, still the same.

    EDIT EDIT: Ok, got it working. Seemingly Stata couldnt get the notion of the usfdi column being non-numerical out of its head, even after I corrected for it. So I justhad to reload everything.


Advertisement