+ Reply to Thread
Results 1 to 2 of 2

Thread: accelerated time model with proportional hazards

  1. #1
    Points: 1,535, Level: 22
    Level completed: 35%, Points required for next Level: 65

    Posts
    18
    Thanks
    3
    Thanked 0 Times in 0 Posts

    accelerated time model with proportional hazards




    $coef
    Estimate SE
    lambda 0.004424690 0.005026449
    gamma 0.888627728 0.108972639
    treatment -0.646997631 0.309927354
    gender -0.701293522 0.392452782
    race 2.105048640 1.015077273
    percentage 0.003336242 0.008991924
    head -0.049555562 0.342919026
    buttocks 0.610699987 0.414709100
    trunk 0.117181285 0.487344728
    upperleg -0.446654667 0.375326008
    lowerleg -0.262843680 0.378571921


    (Intercept) 6.09992 1.2876 4.737 2.16e-06
    treatment 0.72809 0.3606 2.019 4.35e-02
    gender 0.78919 0.4437 1.779 7.53e-02
    race -2.36888 1.1702 -2.024 4.29e-02
    percentage -0.00375 0.0102 -0.369 7.12e-01
    head 0.05577 0.3861 0.144 8.85e-01
    buttocks -0.68724 0.4694 -1.464 1.43e-01
    trunk -0.13187 0.5499 -0.240 8.10e-01
    upperleg 0.50263 0.4236 1.187 2.35e-01
    lowerleg 0.29579 0.4257 0.695 4.87e-01
    Log(scale) 0.11808 0.1226 0.963 3.36e-01
    Scale= 1.13

    Accelerated failure time model
    Log(T)=y= μ+ α^T z+ σW

    Proportional hazards model
    h(x│z)=(γλt^(γ-1) )exp(β^T z)


    γ=1/σ
    λ=exp(-μ/σ)
    β=-α/σ

    Problem if i compute exp(-6.09992/Scale= 1.13) it should be 0.004424690

    But it isnt

    What do i do wrong ?
    What do i do wrong ?

    Is the significance the same for the computed beta's ?
    Last edited by studentbelg; 05-29-2014 at 11:41 AM.

  2. #2
    Points: 1,535, Level: 22
    Level completed: 35%, Points required for next Level: 65

    Posts
    18
    Thanks
    3
    Thanked 0 Times in 0 Posts

    Re: accelerated time model with proportional hazards


    FOUND IT



    Statistics you so fun

+ Reply to 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