Announcement

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

  • Running DEA (data envelopment analysis) with a large sample

    Dear all,
    I have an unbalanced panel of 75 dmus and a span of 27 years (1990-2016) and would like to see how efficiency changes over time. I assigned each entry a number, so effectively Stata 15 sees it as 905 dmus in one year. There are 3 inputs and 2 outouts, so the commands was the following:
    dea i1 i2 i3 = o1 o2, stage (1)
    It took almost an hour and a half to run but now the results are off screen and no way of scrolling further up to see them. Is there any way I can retrieve them? Or run it again but to have the results?
    Thank you in advance.

  • #2
    Welcome to Statalist.

    I don't see any alternative to rerunning the command.

    Check the output of help dea - does it include the trace option? If so, you can use it to save your output to disk as it runs. If no, you perhaps do not have the latest version.

    You can learn the more general approach of using Stata's log capabilities by reading Chapter 15 "Saving and printing output—log files" in the Stata User's Guide PDF included with your Stata installation and accessible from Stata's Help menu.

    You can increase the amount of output that Stata holds in its scroll buffer for the Results window by changing your scrollbufsize setting.
    Code:
    . display c(scrollbufsize)
    512000
    
    . set scrollbufsize 1000000
    (set scrollbufsize will take effect the next time you launch Stata)
    For details see the output of help scrollbufsize.

    Comment


    • #3
      Thank you, William!
      Did not find a trace option but there is an option to save it to a .dta file the next time it is run.

      Comment


      • #4
        Originally posted by Valentina Demchuk View Post
        Thank you, William!
        Did not find a trace option but there is an option to save it to a .dta file the next time it is run.
        Hi Valentina, I have the same problem as you.
        Have you run the results yet?

        Comment

        Working...
        X