|
|
|
||||
|
|
|||||
Table 14.1, page 525.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/rearrest, clear
generate event = ~censor
stset months, failure(event)
failure event: event ~= 0 & event ~= .
obs. time interval: (0, months]
exit on or before: failure
------------------------------------------------------------------------------
194 total obs.
0 exclusions
------------------------------------------------------------------------------
194 obs. remaining, representing
106 failures in single record/single failure data
2678.456 total analysis time at risk, at risk from t = 0
earliest observed entry t = 0
last observed exit t = 36
/* Model A */
stcox personal, nohr
Cox regression -- Breslow method for ties
No. of subjects = 194 Number of obs = 194
No. of failures = 106
Time at risk = 2678.455851
LR chi2(1) = 5.32
Log likelihood = -492.08392 Prob > chi2 = 0.0210
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
personal | .4786654 .202482 2.36 0.018 .0818079 .8755229
------------------------------------------------------------------------------
test personal /* wald test */
( 1) personal = 0.0
chi2( 1) = 5.59
Prob > chi2 = 0.0181
/* Model B */
stcox property, nohr
Cox regression -- Breslow method for ties
No. of subjects = 194 Number of obs = 194
No. of failures = 106
Time at risk = 2678.455851
LR chi2(1) = 16.19
Log likelihood = -486.64864 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
property | 1.194205 .3492838 3.42 0.001 .5096218 1.878789
------------------------------------------------------------------------------
test property /* wald test */
( 1) property = 0.0
chi2( 1) = 11.69
Prob > chi2 = 0.0006
/* Model C */
stcox cage, nohr
Cox regression -- Breslow method for ties
No. of subjects = 194 Number of obs = 194
No. of failures = 106
Time at risk = 2678.455851
LR chi2(1) = 22.93
Log likelihood = -483.27909 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
cage | -.0680667 .0156264 -4.36 0.000 -.0986939 -.0374394
------------------------------------------------------------------------------
test cage /* wald test */
( 1) cage = 0.0
chi2( 1) = 18.97
Prob > chi2 = 0.0000
/* Model D */
stcox personal property cage, nohr
Cox regression -- Breslow method for ties
No. of subjects = 194 Number of obs = 194
No. of failures = 106
Time at risk = 2678.455851
LR chi2(3) = 38.91
Log likelihood = -475.29169 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
personal | .5686725 .2051984 2.77 0.006 .166491 .970854
property | .9354007 .3508835 2.67 0.008 .2476817 1.62312
cage | -.0666248 .0167731 -3.97 0.000 -.0994995 -.0337501
------------------------------------------------------------------------------
test personal /* wald test */
( 1) personal = 0.0
chi2( 1) = 7.68
Prob > chi2 = 0.0056
test property /* wald test */
( 1) property = 0.0
chi2( 1) = 7.11
Prob > chi2 = 0.0077
test cage /* wald test */
( 1) cage = 0.0
chi2( 1) = 15.78
Prob > chi2 = 0.0001
/* Likelihood ratio tests */
lrtest, saving(0)
quietly stcox property cage, nohr /* testing personal */
lrtest
Cox: likelihood-ratio test chi2(1) = 7.27
Prob > chi2 = 0.0070
quietly stcox personal cage, nohr /* testing property */
lrtest
Cox: likelihood-ratio test chi2(1) = 9.14
Prob > chi2 = 0.0025
quietly stcox personal property, nohr /* testing cage */
lrtest
Cox: likelihood-ratio test chi2(1) = 18.28
Prob > chi2 = 0.0000
Table 14.2, page 533.Use predict after stcox personal property cage.
Note: Observations are in numerical id order, not the order shown in the book.
predict risk
list id personal property cage risk months censor if inlist(id,22,8,187,26,5,130,106,33)
id personal property cage risk months censor
5. 5 1 1 -7.164588 7.25298 .2956879 0
8. 8 1 1 22.45074 1.008332 .62423 1
22. 22 0 0 .2577249 .9829757 1.708419 1
26. 26 0 1 -7.301481 4.144825 2.365503 0
33. 33 1 0 27.06129 .2910442 2.792608 1
106. 106 0 0 16.20297 .3397587 11.6961 0
130. 130 0 1 22.39051 .5732906 15.96715 1
187. 187 1 0 -7.200181 2.853034 36 1
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