Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: -predictnl- after -streg-


From   "Feiveson, Alan H. (JSC-SK311)" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   st: -predictnl- after -streg-
Date   Wed, 11 Jul 2012 09:53:27 -0500

Following Steve Samuels' suggestion, I tried -nlpredict- after running -streg-, but I get an error because the estimates of ancillary parameters (kappa and sigma) are involved. Apparently you still have to write out the "complicated" function, whether you use -nlcom- or -nlpredict- (see Steve's second posting on this). 

Too good to be true.


Al Feiveson

. streg urplan,dist(gamma)

         failure _d:  fail
   analysis time _t:  S
  exit on or before:  time .
                 id:  id


Gamma regression -- accelerated failure-time form 

No. of subjects =          213                     Number of obs   =     11850
No. of failures =           24
Time at risk    =        11850
                                                   LR chi2(1)      =     16.19
Log likelihood  =   -63.468446                     Prob > chi2     =    0.0001

------------------------------------------------------------------------------
          _t |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
      urplan |  -1.180329   .2737132    -4.31   0.000    -1.716797   -.6438605
       _cons |   6.323396   .4125432    15.33   0.000     5.514826    7.131966
-------------+----------------------------------------------------------------
     /ln_sig |   .1162692   .3291606     0.35   0.724    -.5288737    .7614121
      /kappa |  -.3319299   1.034859    -0.32   0.748    -2.360216    1.696356
-------------+----------------------------------------------------------------
       sigma |   1.123298   .3697455                      .5892683    2.141298
------------------------------------------------------------------------------

. predictnl B = predict(surv),ci(Blow Bupp)
expression is a function of possibly stochastic quantities other than e(b)
r(498);



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index