SPSS Statistics

 View Only

  • 1.  error messages Mixed Model Analysis

    Posted 24 days ago

    Hello,

    I am using SPSS's Mixed Model Analysis, but in the output, I could not see any result from "Estimates of Covariance Parameters". 

    Here are the error messages below.

    1. Warnings: The final Hessian matrix is not positive definite although all convergence criteria are satisfied. The MIXED procedure continues despite this warning. Validity of subsequent results cannot be ascertained.
    2. Estimates of Covariance Parameters: This covariance parameter is redundant. The test statistic and confidence interval cannot be computed.

    Could someone please let me know the meaning and how can I avoid these messages?

    Thank you



    ------------------------------
    Takka Sun
    ------------------------------


  • 2.  RE: error messages Mixed Model Analysis

    Posted 23 days ago
    Edited by David Dwyer 23 days ago

    Hi @Takka Sun,

    Here is an article from our knowledge base related to your issue:  
    https://www.ibm.com/support/pages/final-hessian-matrix-not-positive-definite-or-failure-converge-warning



    ------------------------------
    David Dwyer
    SPSS Technical Support
    IBM Software
    ------------------------------

    Original Message


  • 3.  RE: error messages Mixed Model Analysis

    Posted 23 days ago

    Hello @David Dwyer,

    Thank you for your information.

    I tried changing the criteria, but the results were the same.

    I would like to know why these results are being displayed and if there might be other causes.

    Also, are these related to this message(result) below?

    • Estimates of Covariance Parameters: This covariance parameter is redundant. The test statistic and confidence interval cannot be computed.

    I would greatly appreciate it if you could provide an answer.

    Thanks,



    ------------------------------
    Takka Sun
    ------------------------------

    Original Message