ASA Connect

 View Only

  • 1.  intraclass correlation

    Posted 10-07-2015 05:26
    Dear colleagues
    I have stata12 and I have analyzed the data using Mixed-Linear models and after getting the results I didn't find the intraclass correlation value with its confidence internal. That's why I have made post estimation using the command (estat icc) but unfortunately I got the below error. I am waiting for your suggestions with my sincere thanks in advance for your kind help.
     Keywords:  r(321)
              Search:  (1) Official help files, FAQs, Examples, SJs, and STBs
    Search of official help files, FAQs, Examples, SJs, and STBs
    [P]     error . . . . . . . . . . . . . . . . . . . . . . . .  Return code 321
            requested action not valid after most recent estimation command;
            This message can be produced by predict or test and indicates
            that the requested action cannot be performed.
    (end of search)
     
    Dr. Faisal G. Khamis, CStat CSci
    Associate Professor of Statistics
    Accredited Professional Statistician, ASA,USA
    Expert at League of Arab States, Egypt
    Consultant in Statistics, RSS, UK
    Fellow of Royal Statistical Society, UK
    Editor-in-chief of International Journal of Statistics and Analysis, India
    College of Business Administration,
    Al Ain University of Science and Technology
    UAE


  • 2.  RE: intraclass correlation

    Posted 10-08-2015 07:41
    As asked your question is not answerable; I suggest you join the
    Statalist forum (http://www.statalist.org/), read the FAQ and then post
    a revised question (showing at least exactly what you entered as a
    command (or set of commands since my guess is that there was something
    between your estimation of the model and your entering "estat icc"))

    Rich


    Original Message


  • 3.  RE: intraclass correlation

    Posted 10-12-2015 09:54

    Dear Dr. Rich

    Thank you so much for your kind answer. I am sorry if my question was ambiguous as I didn’t provide more information. I am studying the rate of such variable in 12 governorates over 14 years. Means that I 14 observations of this variable across 12 clusters. The command I have used is:

    xtmixed divorce year crime, || id: year crime, covariance(independent)

    After I get the results from this command I have used the command estat icc

    Is these information enough? Thank you so much in advance



    ------------------------------
    [Faisal] [Khamis AL-Shamari]
    [Associate Prof.]
    [Al Ain University of Science and Technology]
    ------------------------------


    Original Message