SPSS Statistics

SPSS Statistics

Your hub for statistical analysis, data management, and data documentation. Connect, learn, and share with your peers! 

 View Only
Back to discussions
Expand all | Collapse all

model won't converge unless I standardize variable or divide it by 2

  • 1.  model won't converge unless I standardize variable or divide it by 2

    Posted Tue March 02, 2021 08:11 PM

    I am running a mixed model that will not converge with a raw Time variable (1-8), which is included as fixed and random effect. What is strange and troubling is that the exact same model will converge if I either standardize the Time variable or divide it by 2 (so it is .5-4). Moreover, the exact some model that will not converge for me (raw Time variable) also will not converge for some of my students but WILL converge for some of my other students. The model will not converge for me using SPSS 26 or SPSS 27. Here is the syntax:

    MIXED nwms WITH Time ms_cn bdi_cn

    /CRITERIA=CIN(95) MXITER(10000) MXSTEP(100) SCORING(1) SINGULAR(0.000000000001) HCONVERGE(0,

    ABSOLUTE) LCONVERGE(0, ABSOLUTE) PCONVERGE(0.000001, ABSOLUTE)

    /FIXED= Time ms_cn bdi_cn | SSTYPE(3)

    /METHOD=REML

    /PRINT= SOLUTION TESTCOV

    /RANDOM= intercept Time ms_cn bdi_cn | SUBJECT(IDnum) COVTYPE(UN).






    #SPSSStatistics
    #Support
    #SupportMigration


  • 2.  RE: model won't converge unless I standardize variable or divide it by 2

    Posted Wed March 10, 2021 10:04 AM

    Support Member​ 

    Dear SPSS User,

    if you have a statistical or coding question please use the IBM Data Science Community for SPSS Statistics instead of using this forum .

    Please go to the link below and ask your question:

     

    https://community.ibm.com/community/user/datascience/communities/community-home/digestviewer?communitykey=886b6874-0fb1-402c-8243-c70ef8179a99&tab=digestviewer

     

    Thank you.

    regards, IBM SPSS Technical Support






    #SPSSStatistics
    #Support
    #SupportMigration


Related Content