SPSS Statistics

SPSS Statistics

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

 View Only
  • 1.  Error using COMPRISK extension

    Posted Mon April 25, 2022 07:28 AM
    I am trying to perform a competing risk survival analysis in SPSS on MS Windows 10 using the COMPRISK extension command.

    My syntax (generated through the GUI) is

    STATS COMPRISK
    FAILTIME=Month_Last FAILSTATUS=Status FIXEDCOVAR=TXtype FAILCODE=2 CENSORCODE=0
    /OPTIONS MISSING=OMIT PLOT=YES QUANTILES=.25 .50 .75 MAXITER=10 TOL=.000001.

    ... and the error message that I am receiving is

    Error in names(coef) <- c(gtxt("Coefficient"), gtxt("Exp"), gtxt("Std. Error"), :
    'names' attribute [7] must be the same length as the vector [6]
    Calls: Run ... spsspkg.processcmd -> do.call -> docomprisk -> displayresults
    In addition: Warning message:
    In data.frame(ressum$coef, ci) :
    row names were found from a short variable and have been discarded

    Any ideas?

    ------------------------------
    Frank Furter
    ------------------------------

    #SPSSStatistics


  • 2.  RE: Error using COMPRISK extension

    Posted Mon April 25, 2022 08:45 AM
    I suspect that this is a problem with missing data that was addressed in an update.

    Strangely, this site won't let me paste the url for the update into this email, claiming that the page doesn't exist.
    Send me an email (jkpeck@gmail.com), and I'll reply with the url.

    If that doesn't resolve the problem, please send me data and syntax, and I will investigate.  Let me know, too, if the update resolves the problem.


    ------------------------------
    Jon Peck
    ------------------------------



  • 3.  RE: Error using COMPRISK extension

    Posted Tue April 26, 2022 05:45 AM
    Dear Jon:

    Thanks so much for providing the URL, which worked fine. I installed the updated package from the link, and now the COMPRISK procedures works without errors.

    BTW: My understanding of the procedure is that it provides an extension of Cox regression analysis to competing risks (right?). Is there also an SPSS extension procedure for cumulative incidence survival analysis with competing risks (i.e., for determining non-parametric, maximum-likelihood estimates and confidence limits of the probability of failure for a particular cause in the presence of other causes, including group comparisons using Gray's test)?

    ------------------------------
    Frank Furter
    ------------------------------



  • 4.  RE: Error using COMPRISK extension

    Posted Tue April 26, 2022 08:39 AM
    I'm glad that worked out.  Your understanding is correct, but AFAIK, there is no extension for that additional algorithm.

    --