Re: st: Weibull simulation-parametrization

[Antoine Terracol <Antoine.Terracol@univ-paris1.fr>]

n p wrote:

> I would appreciate any other ideas for a Weibull
> simulation or even better a Weibull with frailties
> simulation.
>
> Thanks in advance for any comments.
>
> Nikos Pantazis

Nikos,

the following piece of code will generate weibull random variables (my 
alpha stands for your gamma).
------------------------------------------
cap prog drop draweib
program define draweib
syntax newvarlist  [if] [in] , LAmbda(string) ALpha(string) [double]
tokenize "`varlist'"
while "`1'"!="" {
tempname vlambda
tempvar `vlambda'
gen `vlambda' = exp(ln(`lambda')/`alpha')
g `double' `1' = ((log(1/uniform()))^(1/`alpha'))/`vlambda' `if' `in'
mac shift
}
end
------------------------------------------

You can also use it to simulate durations that depend on some 
independant variables as in the output below.

. clear

. set obs 1000
obs was 0, now 1000

. drawnorm x1 x2

. draweib time, lambda(exp(1+x1+x2)) alpha(0.6)

. stset time

     failure event:  (assumed to fail at time=time)
obs. time interval:  (0, time]
 exit on or before:  failure

------------------------------------------------------------------------------
     1000  total obs.
        0  exclusions
------------------------------------------------------------------------------
     1000  obs. remaining, representing
     1000  failures in single record/single failure data
  4534.04  total analysis time at risk, at risk from t =         0
                             earliest observed entry t =         0
                                  last observed exit t =  1827.592

. streg x1 x2, dist(weib) nohr nolog

         failure _d:  1 (meaning all fail)
   analysis time _t:  time

Weibull regression -- log relative-hazard form

No. of subjects =         1000                     Number of obs   = 
  1000
No. of failures =         1000
Time at risk    =   4534.03956
                                                   LR chi2(2)      = 
1069.97
Log likelihood  =   -2084.7593                     Prob > chi2     = 
0.0000

------------------------------------------------------------------------------
          _t |      Coef.   Std. Err.      z    P>|z|     [95% Conf. 
Interval]
-------------+----------------------------------------------------------------
          x1 |   1.030093    .041635    24.74   0.000     .9484899 
1.111696
          x2 |   1.012599   .0394268    25.68   0.000     .9353235 
1.089874
       _cons |   1.079722   .0357241    30.22   0.000     1.009704 
1.14974
-------------+----------------------------------------------------------------
       /ln_p |  -.4970151   .0248969   -19.96   0.000    -.5458122 
-.448218
-------------+----------------------------------------------------------------
           p |   .6083438   .0151459                       .579371 
.6387654
         1/p |   1.643807   .0409258                       1.56552 
1.72601
------------------------------------------------------------------------------



Hope this helps,
Antoine.



--
Ce message a subi une analyse antivirus 
par MailScanner ; il est vraisemblablement
sans danger.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/