+ Reply to Thread
Results 1 to 2 of 2

Thread: cox regression

  1. #1
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    cox regression




    This is interested in surival time to an event and deals with time series data inherently, including the use of lagging indicators in some cases. Many of the independent variables can change with time.

    Discussions I have seen do not address concerns with autocorrelated data or bias tied to lagging predictors discussed in the time series literature. For instance test of autocorrelation common in time series are not raised [or have not in the literature I have seen anyhow}.

    Are these not issues, not concerns, with Cox regression? I know the approach uses partial likelihood, perhaps this elminates the issues of autocorrelation and lagged variables.
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

  2. #2
    Fortran must die
    Points: 58,790, Level: 100
    Level completed: 0%, Points required for next Level: 0
    noetsi's Avatar
    Posts
    6,532
    Thanks
    692
    Thanked 915 Times in 874 Posts

    Re: cox regression


    Really dumb question If a hazard ratio interval contains 0 it means you can not be sure that a variable influenced the time to event right? I ask because SAS apparently does not generate test of statistical signficance for simple effects when interaction occurs, but does generate say the 95 CI for those effects.

    So you can specify the impact of var1 at specific levels of var 2 [what I call a simple effect] for COX regression and you will get a ci but you won't get a statistical test of signficance at each level.

    Perhaps you can't actually do statistical test for simple effects - my ignorance is vast...
    "Very few theories have been abandoned because they were found to be invalid on the basis of empirical evidence...." Spanos, 1995

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats