请教各位老师，我在尝试使用Threshold regression时，stata 总是报错“numerical derivatives are approximate
nearby values are missing” （程序如下，其中变量赋值water(0,1), asa(0,1), ix (1,2,3,4), dia (0,1), upc (0,1), med (0,1), ven (0,1), who (0,1,2,3,4), age2 (18-50)）


.......有哪位老师知道发生了什么吗?.....是哪里出了问题......?
谢谢各位老师！

---------------------------------------------------------------------
. xi: stthreg, lny0(water asa ix dia upc med ven  who age2) mu
> (water asa ix dia upc med ven who age2)

         failure _d:  censored2
   analysis time _t:  ex

initial:       log likelihood = -683.54125
alternative:   log likelihood = -587.11961
rescale:       log likelihood = -536.38588
rescale eq:    log likelihood = -257.71142
Iteration 0:   log likelihood = -257.71142  
Iteration 1:   log likelihood = -246.64379  
Iteration 2:   log likelihood = -235.47662  
Iteration 3:   log likelihood = -227.74748  
Iteration 4:   log likelihood = -227.19192  
Iteration 5:   log likelihood =  -226.0596  
Iteration 6:   log likelihood = -225.94916  
Iteration 7:   log likelihood = -225.89445  
Iteration 8:   log likelihood = -225.88972  (backed up)
Iteration 9:   log likelihood = -225.88518  (backed up)
Iteration 10:  log likelihood = -225.88511  (backed up)
numerical derivatives are approximate
nearby values are missing
Iteration 11:  log likelihood = -225.88509  (backed up)
numerical derivatives are approximate
nearby values are missing
Iteration 12:  log likelihood = -225.88508  (backed up)
numerical derivatives are approximate
nearby values are missing
numerical derivatives are approximate
nearby values are missing
Iteration 13:  log likelihood = -225.88508  (backed up)
numerical derivatives are approximate
nearby values are missing
numerical derivatives are approximate
nearby values are missing
Iteration 14:  log likelihood = -225.88508  (backed up)
numerical derivatives are approximate
nearby values are missing
numerical derivatives are approximate
nearby values are missing
Iteration 15:  log likelihood = -225.88508  (not concave)
numerical derivatives are approximate
nearby values are missing
numerical derivatives are approximate
nearby values are missing
Iteration 16:  log likelihood = -225.88508  (not concave)
numerical derivatives are approximate
nearby values are missing
numerical derivatives are approximate
nearby values are missing
Iteration 17:  log likelihood = -225.88508  (not concave)
numerical derivatives are approximate
nearby values are missing
could not calculate numerical derivatives
missing values encountered
r(430);
----------------------------------------------------------------------