Missing SE in cox regression.

Dear All,
I am new to survival analysis using Stata and am trying to estimate a standard cox proportional hazard regression. For some reason though one of the standard errors is not estimated. Here is the dataex:

Code:

* Example generated by -dataex-. To install: ssc install dataex
clear
input long case_number float(event tot type tot_cat male) byte(Black Hispanic Other age2534 age3544 age4554 age55plus edulhs edulcol educol edupg edumiss)
100064 1  0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0
100076 1  0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 0
100079 1  1 1 1 0 0 0 0 1 0 0 0 0 1 0 0 0
100092 1  0 0 0 1 1 0 0 0 0 1 0 0 0 1 0 0
100095 1  0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0
100124 1  0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0
100127 1  5 1 3 0 0 0 0 0 1 0 0 1 0 0 0 0
100154 1  0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0
100181 1  0 0 0 1 1 0 0 0 0 1 0 0 0 1 0 0
100191 1  4 1 2 1 0 0 0 0 0 1 0 1 0 0 0 0
100195 1  0 0 0 1 0 0 0 0 1 0 0 1 0 0 0 0
100197 1  6 1 3 0 0 0 0 0 0 1 0 1 0 0 0 0
100210 1  0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0
100227 1  0 0 0 1 1 0 0 0 0 1 0 1 0 0 0 0
100256 1  0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0
100264 1  0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
100269 1  2 1 1 0 0 0 0 1 0 0 0 1 0 0 0 0
100289 1  0 0 0 0 1 0 0 0 0 1 0 0 1 0 0 0
100302 1  2 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0
100319 1  0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0
100330 1  1 1 1 1 0 0 0 1 0 0 0 0 1 0 0 0
100332 1  1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0
100337 1  1 1 1 1 0 0 0 1 0 0 0 0 0 1 0 0
100338 1  1 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0
100339 1  0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0
100343 1  0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0
100344 1  0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 0
100346 1  0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0
100357 1  3 1 2 0 0 0 0 0 0 0 1 0 0 1 0 0
100359 1  1 1 1 0 0 0 0 0 0 0 1 0 0 1 0 0
100372 1  1 1 1 0 0 0 0 1 0 0 0 0 0 0 0 1
100377 1  0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0
100386 1  1 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0
100402 1  0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0
100408 1  0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0
100417 1  1 1 1 0 1 0 0 0 0 1 0 0 0 1 0 0
100423 1  0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0
100432 1  0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0
100438 1  0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0
100447 1  0 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0
100455 1  0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
100461 1  0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0
100480 1  0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 0
100487 1  9 1 4 1 0 0 0 0 0 0 0 0 0 0 0 0
100502 1  0 0 0 1 0 0 0 1 0 0 0 1 0 0 0 0
100507 1  4 1 2 1 0 0 0 1 0 0 0 0 1 0 0 0
100530 1  3 1 2 0 0 0 0 0 0 1 0 0 0 0 0 0
100540 1  1 1 1 1 0 0 0 1 0 0 0 0 1 0 0 0
100544 1  0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0
100562 1  0 0 0 1 0 0 0 0 0 1 0 1 0 0 0 0
100563 1  3 1 2 1 0 0 0 0 0 0 0 1 0 0 0 0
100568 1  0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0
100580 1  0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1
100595 1  1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0
100605 1  0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 0
100618 1  0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0
100636 1  2 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0
100638 1  0 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0
100644 1  0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
100652 1  1 1 1 1 0 0 0 1 0 0 0 0 0 0 0 0
100678 1  1 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0
100691 1  0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
100696 1  0 0 0 1 1 0 0 0 0 0 1 0 0 0 0 0
100710 1  5 1 3 1 0 0 0 0 0 1 0 0 0 0 0 0
100713 1  1 1 1 1 0 0 0 0 1 0 0 0 0 0 1 0
100715 1  2 1 1 1 0 0 0 0 1 0 0 1 0 0 0 0
100732 1  3 1 2 1 0 0 0 0 0 1 0 0 0 0 0 0
100734 1  0 0 0 1 1 0 0 0 0 1 0 0 0 1 0 0
100737 1  0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0
100761 1  1 1 1 1 0 0 0 0 1 0 0 0 0 0 0 0
100772 1  0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0
100794 1  0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
100802 1  1 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0
100840 1  2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0
100857 1  2 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0
100861 1  3 1 2 0 0 0 0 0 1 0 0 0 0 0 0 0
100862 1  2 1 1 1 0 0 0 1 0 0 0 0 0 0 0 0
100864 1  0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0
100872 1  2 1 1 0 0 0 0 1 0 0 0 0 0 0 0 1
100882 1  0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0
100884 1  1 1 1 0 1 0 0 0 0 0 1 0 0 0 0 1
100891 1  5 1 3 1 0 0 0 0 1 0 0 0 0 0 0 0
100896 1  1 1 1 1 0 0 0 1 0 0 0 1 0 0 0 0
100908 1  0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0
100928 1  3 1 2 1 0 0 0 0 1 0 0 0 0 0 0 1
100933 1  0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
100954 1  0 0 0 1 1 0 0 0 1 0 0 1 0 0 0 0
100965 1  1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0
100985 1  3 1 2 0 0 0 0 0 0 1 0 0 0 0 0 0
100996 1  1 1 1 0 1 0 0 0 1 0 0 0 0 0 0 1
101007 1  3 1 2 1 0 0 0 1 0 0 0 0 0 0 0 0
101013 1  0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
101024 1  2 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0
101029 1  0 0 0 1 1 0 0 1 0 0 0 1 0 0 0 0
101071 1  0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0
101082 1  2 1 1 1 0 0 0 1 0 0 0 0 1 0 0 0
101084 1  3 1 2 1 1 0 0 1 0 0 0 0 1 0 0 0
101085 1  0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0
101091 1 12 1 4 0 0 0 0 0 0 0 1 0 1 0 0 0
101112 1  0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0
end

'tot' is the continuous treatment variable.

tot | Freq. Percent Cum.
------------+-----------------------------------
0 | 871 38.66 38.66
1 | 542 24.06 62.72
2 | 316 14.03 76.74
3 | 189 8.39 85.13
4 | 114 5.06 90.19
5 | 65 2.89 93.08
6 | 41 1.82 94.90
7 | 38 1.69 96.58
8 | 27 1.20 97.78
9 | 8 0.36 98.14
10 | 5 0.22 98.36
11 | 9 0.40 98.76
12 | 6 0.27 99.02
13 | 3 0.13 99.16
14 | 5 0.22 99.38
15 | 2 0.09 99.47
16 | 4 0.18 99.64
17 | 2 0.09 99.73
19 | 2 0.09 99.82
20 | 1 0.04 99.87
21 | 1 0.04 99.91
22 | 1 0.04 99.96
38 | 1 0.04 100.00
------------+-----------------------------------
Total | 2,253 100.00

'type' is a binary recode of tot:

Code:

gen type=(tot>0)

And, 'tot_cat' is a categorical recode:

Code:

recode tot (1/2=1) (3/4=2) (5/8=3) (9/38=4), gen(tot_cat)recotype tot_cat

I ran the following:

Code:

. stset dur, id(case_number) failure(event==1) scale(1)

                id:  case_number
     failure event:  event == 1
obs. time interval:  (dur[_n-1], dur]
 exit on or before:  failure

------------------------------------------------------------------------------
      2,253  total observations
        281  observations end on or before enter()
------------------------------------------------------------------------------
      1,972  observations remaining, representing
      1,972  subjects
      1,972  failures in single-failure-per-subject data
     78,376  total analysis time at risk and under observation
                                                at risk from t =         0
                                     earliest observed entry t =         0
                                          last observed exit t =        52

. 
end of do-file

. do "C:\Users\Sumedha\AppData\Local\Temp\STD3208_000000.tmp"

. stcox tot male Black Hispanic Other age2534 age3544 age4554 age55plus edulhs edulcol educol edupg edumiss empstud, no
> hr

         failure _d:  event == 1
   analysis time _t:  dur
                 id:  case_number

Iteration 0:   log likelihood = -13445.298
Iteration 1:   log likelihood = -13325.559
Iteration 2:   log likelihood =  -13316.82
Iteration 3:   log likelihood = -13316.713
Iteration 4:   log likelihood = -13316.713
Refining estimates:
Iteration 0:   log likelihood = -13316.713

Cox regression -- Breslow method for ties

No. of subjects =        1,918                  Number of obs    =       1,918
No. of failures =        1,918
Time at risk    =        76464
                                                LR chi2(15)      =      257.17
Log likelihood  =   -13316.713                  Prob > chi2      =      0.0000

------------------------------------------------------------------------------
          _t |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
         tot |   .0960208   .0049154    19.53   0.000     .0863868    .1056548
        male |  -.0705655   .0489316    -1.44   0.149    -.1664697    .0253387
       Black |  -.0197382   .0607501    -0.32   0.745    -.1388062    .0993299
    Hispanic |  -.0635071    .204539    -0.31   0.756    -.4643962     .337382
       Other |   -.147056   .1756267    -0.84   0.402     -.491278     .197166
     age2534 |   .0619619   .0853876     0.73   0.468    -.1053947    .2293185
     age3544 |  -.0508297   .0882918    -0.58   0.565    -.2238785    .1222192
     age4554 |  -.1089686   .0886928    -1.23   0.219    -.2828033     .064866
   age55plus |   -.120924   .0961707    -1.26   0.209    -.3094151     .067567
      edulhs |  -.0368119   .0586314    -0.63   0.530    -.1517273    .0781035
     edulcol |  -.0177244   .0746792    -0.24   0.812     -.164093    .1286442
      educol |  -.0496405   .0942559    -0.53   0.598    -.2343786    .1350976
       edupg |  -.0267162   .2407701    -0.11   0.912     -.498617    .4451846
     edumiss |  -.0141879   .0757348    -0.19   0.851    -.1626254    .1342496
     empstud |  -.1566901   .0507535    -3.09   0.002    -.2561652    -.057215
------------------------------------------------------------------------------

. stcox i.tot_cat male Black Hispanic Other age2534 age3544 age4554 age55plus edulhs edulcol educol edupg edumiss, nohr

         failure _d:  event == 1
   analysis time _t:  dur
                 id:  case_number

Iteration 0:   log likelihood = -13871.757
Iteration 1:   log likelihood = -12734.904
Iteration 2:   log likelihood = -12569.655
Iteration 3:   log likelihood = -12534.636
Iteration 4:   log likelihood = -12527.617
Iteration 5:   log likelihood = -12525.127
Iteration 6:   log likelihood =  -12524.22
Iteration 7:   log likelihood = -12523.888
Iteration 8:   log likelihood = -12523.766
Iteration 9:   log likelihood = -12523.721
Iteration 10:  log likelihood = -12523.704
Iteration 11:  log likelihood = -12523.698
Iteration 12:  log likelihood = -12523.696
Iteration 13:  log likelihood = -12523.695
Iteration 14:  log likelihood = -12523.695
Iteration 15:  log likelihood = -12523.695
Iteration 16:  log likelihood = -12523.695  (backed up)
Iteration 17:  log likelihood = -12523.695
Iteration 18:  log likelihood = -12523.694
Iteration 19:  log likelihood = -12523.694
Iteration 20:  log likelihood = -12523.693
Iteration 21:  log likelihood = -12523.696  (backed up)
Refining estimates:
Iteration 0:   log likelihood = -12523.695
Iteration 1:   log likelihood = -12523.695
Iteration 2:   log likelihood = -12523.695
Iteration 3:   log likelihood = -12523.695
Iteration 4:   log likelihood = -12523.695
Iteration 5:   log likelihood = -12523.695
Iteration 6:   log likelihood = -12523.695

Cox regression -- Breslow method for ties

No. of subjects =        1,972                  Number of obs    =       1,972
No. of failures =        1,972
Time at risk    =        78376
                                                LR chi2(16)      =     2696.13
Log likelihood  =   -12523.695                  Prob > chi2      =      0.0000

------------------------------------------------------------------------------
          _t |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     tot_cat |
          1  |   26.23135   .1499912   174.89   0.000     25.93737    26.52533
          2  |   25.82186   .1547281   166.89   0.000      25.5186    26.12512
          3  |   25.34002   .1617857   156.63   0.000     25.02293    25.65712
          4  |   25.09946          .        .       .            .           .
             |
        male |   .0789911   .0483001     1.64   0.102    -.0156753    .1736575
       Black |   -.037212   .0600565    -0.62   0.536    -.1549206    .0804966
    Hispanic |   .0511079   .2037287     0.25   0.802    -.3481929    .4504087
       Other |   .0108578   .1729652     0.06   0.950    -.3281478    .3498634
     age2534 |  -.1162275    .084623    -1.37   0.170    -.2820855    .0496304
     age3544 |  -.0931516   .0868272    -1.07   0.283    -.2633297    .0770265
     age4554 |  -.0696997    .087131    -0.80   0.424    -.2404734     .101074
   age55plus |  -.0783454   .0943127    -0.83   0.406    -.2631948    .1065041
      edulhs |  -.0138597   .0576137    -0.24   0.810    -.1267805    .0990611
     edulcol |    .042249   .0745802     0.57   0.571    -.1039254    .1884234
      educol |    .049197   .0942711     0.52   0.602     -.135571    .2339649
       edupg |   -.057499   .2402637    -0.24   0.811    -.5284072    .4134093
     edumiss |   .0892369   .0683832     1.30   0.192    -.0447918    .2232656
------------------------------------------------------------------------------

. stcox type male Black Hispanic Other age2534 age3544 age4554 age55plus edulhs edulcol educol edupg edumiss empstud, n
> ohr

         failure _d:  event == 1
   analysis time _t:  dur
                 id:  case_number

Iteration 0:   log likelihood = -13445.298
Iteration 1:   log likelihood = -12336.382
Iteration 2:   log likelihood = -12247.619
Iteration 3:   log likelihood = -12220.996
Iteration 4:   log likelihood = -12211.957
Iteration 5:   log likelihood = -12208.731
Iteration 6:   log likelihood = -12207.558
Iteration 7:   log likelihood = -12207.128
Iteration 8:   log likelihood =  -12206.97
Iteration 9:   log likelihood = -12206.912
Iteration 10:  log likelihood =  -12206.89
Iteration 11:  log likelihood = -12206.883
Iteration 12:  log likelihood =  -12206.88
Iteration 13:  log likelihood = -12206.879
Iteration 14:  log likelihood = -12206.878
Iteration 15:  log likelihood = -12206.878
Iteration 16:  log likelihood = -12206.878
Iteration 17:  log likelihood = -12206.878
Iteration 18:  log likelihood = -12206.878
Iteration 19:  log likelihood = -12206.878
Refining estimates:
Iteration 0:   log likelihood = -12206.878
Iteration 1:   log likelihood = -12206.878
Iteration 2:   log likelihood = -12206.878
Iteration 3:   log likelihood = -12206.878
Iteration 4:   log likelihood = -12206.878
Iteration 5:   log likelihood = -12206.878
Iteration 6:   log likelihood = -12206.878
Iteration 7:   log likelihood = -12206.878
Iteration 8:   log likelihood = -12206.878
Iteration 9:   log likelihood = -12206.878
Iteration 10:  log likelihood = -12206.878
Iteration 11:  log likelihood = -12206.878
Iteration 12:  log likelihood = -12206.878
Iteration 13:  log likelihood = -12206.878
Iteration 14:  log likelihood = -12206.878
Iteration 15:  log likelihood = -12206.878
Iteration 16:  log likelihood = -12206.878
Iteration 17:  log likelihood = -12206.878
Iteration 18:  log likelihood = -12206.878
Iteration 19:  log likelihood = -12206.878
Iteration 20:  log likelihood = -12206.878
Iteration 21:  log likelihood = -12206.878
Iteration 22:  log likelihood = -12206.878
Iteration 23:  log likelihood = -12206.878

Cox regression -- Breslow method for ties

No. of subjects =        1,918                  Number of obs    =       1,918
No. of failures =        1,918
Time at risk    =        76464
                                                LR chi2(14)      =     2476.84
Log likelihood  =   -12206.878                  Prob > chi2      =      0.0000

------------------------------------------------------------------------------
          _t |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        type |   41.42786          .        .       .            .           .
        male |   .0959389    .048961     1.96   0.050    -.0000229    .1919007
       Black |  -.0421008   .0608911    -0.69   0.489    -.1614451    .0772436
    Hispanic |    .067613   .2037077     0.33   0.740    -.3316468    .4668728
       Other |  -.0025301   .1757478    -0.01   0.989    -.3469895    .3419294
     age2534 |  -.0763014   .0853566    -0.89   0.371    -.2435973    .0909945
     age3544 |  -.0684269   .0875539    -0.78   0.434    -.2400293    .1031756
     age4554 |  -.0872988   .0882541    -0.99   0.323    -.2602736     .085676
   age55plus |  -.0830078   .0955541    -0.87   0.385    -.2702904    .1042747
      edulhs |  -.0107608   .0583419    -0.18   0.854    -.1251088    .1035873
     edulcol |   .0899122   .0747195     1.20   0.229    -.0565353    .2363597
      educol |   .1073924   .0945303     1.14   0.256    -.0778835    .2926684
       edupg |   .0172314   .2404049     0.07   0.943    -.4539536    .4884164
     edumiss |   .0313791    .076609     0.41   0.682    -.1187718    .1815301
     empstud |  -.0228335   .0514697    -0.44   0.657    -.1237123    .0780453
------------------------------------------------------------------------------

I am not sure why the SE is not getting estimated, except in the case of the continuous covariate 'tot'. I will be grateful for any suggestions you may have.
Sincerely,
Sumedha.

BJ Data Tech Solution

Home / Data Cleaning / Data management / Data Processing / Missing SE in cox regression.
Missing SE in cox regression.

0 Response to Missing SE in cox regression.

Post a Comment

Home / Data Cleaning / Data management / Data Processing / Missing SE in cox regression. Missing SE in cox regression.

0 Response to Missing SE in cox regression.

Post a Comment

Home / Data Cleaning / Data management / Data Processing / Missing SE in cox regression.
Missing SE in cox regression.