Help the Stat Consulting Group by
giving a gift
Loading
|
|
|
||||
|
Help the Stat Consulting Group by
giving a gift
| |||||
|
Loading
|
|||||
Table 3.1 on page 78 using the actg320 dataset.
use http://www.ats.ucla.edu/stat/examples/asa2/actg320, clear
stset time, fail(censor)
stcox tx, nohr
failure _d: censor
analysis time _t: time
Iteration 0: log likelihood = -658.46549
Iteration 1: log likelihood = -653.12286
Iteration 2: log likelihood = -653.11789
Iteration 3: log likelihood = -653.11789
Refining estimates:
Iteration 0: log likelihood = -653.11789
Cox regression -- Breslow method for ties
No. of subjects = 1151 Number of obs = 1151
No. of failures = 96
Time at risk = 264941
LR chi2(1) = 10.70
Log likelihood = -653.11789 Prob > chi2 = 0.0011
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
tx | -.6843186 .2149187 -3.18 0.001 -1.105551 -.2630857
------------------------------------------------------------------------------
Table 3.2 on page 83 continuing to use the actg320 dataset.
stcox tx age sex cd4 priorzdv, nohr
failure _d: censor
analysis time _t: time
Iteration 0: log likelihood = -658.46549
Iteration 1: log likelihood = -622.46228
Iteration 2: log likelihood = -618.94567
Iteration 3: log likelihood = -618.82581
Iteration 4: log likelihood = -618.8256
Refining estimates:
Iteration 0: log likelihood = -618.8256
Cox regression -- Breslow method for ties
No. of subjects = 1151 Number of obs = 1151
No. of failures = 96
Time at risk = 264941
LR chi2(5) = 79.28
Log likelihood = -618.8256 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
tx | -.6589906 .2152889 -3.06 0.002 -1.080949 -.2370322
age | .0283562 .0112658 2.52 0.012 .0062756 .0504368
sex | .0972738 .2841181 0.34 0.732 -.4595874 .654135
cd4 | -.0165767 .0025453 -6.51 0.000 -.0215654 -.0115879
priorzdv | -.0002931 .0036913 -0.08 0.937 -.007528 .0069417
------------------------------------------------------------------------------
Table 3.3 on page 84 continuing to use the actg320 dataset.
stcox tx age cd4, nohr
failure _d: censor
analysis time _t: time
Iteration 0: log likelihood = -658.46549
Iteration 1: log likelihood = -622.50294
Iteration 2: log likelihood = -619.00548
Iteration 3: log likelihood = -618.8856
Iteration 4: log likelihood = -618.88539
Refining estimates:
Iteration 0: log likelihood = -618.88539
Cox regression -- Breslow method for ties
No. of subjects = 1151 Number of obs = 1151
No. of failures = 96
Time at risk = 264941
LR chi2(3) = 79.16
Log likelihood = -618.88539 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
tx | -.6587076 .2150446 -3.06 0.002 -1.080187 -.237228
age | .0277729 .0111474 2.49 0.013 .0059243 .0496215
cd4 | -.0165606 .0025391 -6.52 0.000 -.0215372 -.011584
------------------------------------------------------------------------------
Table 3.4 on page 87 using the whas100 dataset. We convert time to quarterly to increase the number of ties.
use http://www.ats.ucla.edu/stat/examples/asa2/whas100, clear
generate time=foltime/30.44 /* divide by days per month */
replace time=round(time/3) /* divide by months per quarter */
replace time=.5 if time==0 /* event can't occur at time zero */
stset time, fail(folstatus)
/* exact ties */
stcox bmi gender, nohr exactp
failure _d: folstatus
analysis time _t: time
Iteration 0: log likelihood = -183.56906
Iteration 1: log likelihood = -177.82554
Iteration 2: log likelihood = -177.79731
Iteration 3: log likelihood = -177.79731
Refining estimates:
Iteration 0: log likelihood = -177.79731
Cox regression -- exact partial likelihood
No. of subjects = 100 Number of obs = 100
No. of failures = 51
Time at risk = 1648
LR chi2(2) = 11.54
Log likelihood = -177.79731 Prob > chi2 = 0.0031
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
bmi | -.0921293 .0337883 -2.73 0.006 -.1583532 -.0259055
gender | .5391896 .2875519 1.88 0.061 -.0244017 1.102781
------------------------------------------------------------------------------
/* Breslow ties */
stcox bmi gender, nohr
failure _d: folstatus
analysis time _t: time
Iteration 0: log likelihood = -210.71578
Iteration 1: log likelihood = -205.1773
Iteration 2: log likelihood = -205.14533
Iteration 3: log likelihood = -205.14533
Refining estimates:
Iteration 0: log likelihood = -205.14533
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 51
Time at risk = 1648
LR chi2(2) = 11.14
Log likelihood = -205.14533 Prob > chi2 = 0.0038
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
bmi | -.0885065 .0329861 -2.68 0.007 -.1531582 -.0238549
gender | .5181654 .2830296 1.83 0.067 -.0365625 1.072893
------------------------------------------------------------------------------
/* Efron ties */
stcox bmi gender, nohr efron
failure _d: folstatus
analysis time _t: time
Iteration 0: log likelihood = -210.06393
Iteration 1: log likelihood = -204.20163
Iteration 2: log likelihood = -204.16685
Iteration 3: log likelihood = -204.16685
Refining estimates:
Iteration 0: log likelihood = -204.16685
Cox regression -- Efron method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 51
Time at risk = 1648
LR chi2(2) = 11.79
Log likelihood = -204.16685 Prob > chi2 = 0.0027
------------------------------------------------------------------------------
_t | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
bmi | -.0925082 .0334316 -2.77 0.006 -.1580329 -.0269835
gender | .5332732 .282779 1.89 0.059 -.0209634 1.08751
------------------------------------------------------------------------------
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services