|
|
|
||||
|
|
|||||
The data files used for the examples in this text can be downloaded in a zip file from the Wiley FTP website or the Stata Web site. You can then use a program such as WinZip to unzip the data files. If you need assistance getting data into Stata, please see our Stata Class Notes, especially the unit on Entering Data. (NOTE: The *.dat files are the data files, and the *.txt files contain the codebook information.)
use hmohiv, clear stset time, failure(censor)
Table 4.2, page 119.
stcox drug, nohr
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(1) = 10.20
Log likelihood = -294.09639 Prob > chi2 = 0.0014
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
drug | .7792305 .2422582 3.22 0.001 .3044132 1.254048
------------------------------------------------------------------------------
Table 4.3, page 121.
generate agegrp = age
recode agegrp 20/29=1 30/34=2 35/39=3 40/54=4
tab agegrp, gen(age)
agegrp | Freq. Percent Cum.
------------+-----------------------------------
1 | 12 12.00 12.00
2 | 34 34.00 46.00
3 | 25 25.00 71.00
4 | 29 29.00 100.00
------------+-----------------------------------
Total | 100 100.00
Table 4.4, page 122.
stcox age2 age3 age4, nohr
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(3) = 19.56
Log likelihood = -289.41333 Prob > chi2 = 0.0002
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age2 | 1.196962 .4510352 2.65 0.008 .3129497 2.080975
age3 | 1.313291 .4588966 2.86 0.004 .4138706 2.212712
age4 | 1.8604 .469308 3.96 0.000 .9405736 2.780227
------------------------------------------------------------------------------
Table 4.5, page 123.
stcox age2 age3 age4
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(3) = 19.56
Log likelihood = -289.41333 Prob > chi2 = 0.0002
------------------------------------------------------------------------------
_t |
_d | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age2 | 3.310047 1.492948 2.65 0.008 1.367453 8.012279
age3 | 3.718392 1.706358 2.86 0.004 1.512661 9.140474
age4 | 6.42631 3.015919 3.96 0.000 2.56145 16.12269
------------------------------------------------------------------------------
Table 4.4, page 125.
matrix list e(V)
symmetric e(V)[3,3]
age2 age3 age4
age2 .20343276
age3 .16365896 .2105861
age4 .17047514 .16661653 .22025002
Table 4.7, page 127.
replace age2 = age2 - age1
replace age3 = age3 - age1
replace age4 = age4 - age1
stcox age2 age3 age4, nohr
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(3) = 19.56
Log likelihood = -289.41333 Prob > chi2 = 0.0002
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age2 | .1042989 .1920546 0.54 0.587 -.2721212 .480719
age3 | .2206278 .2058906 1.07 0.284 -.1829103 .624166
age4 | .7677369 .2093199 3.67 0.000 .3574775 1.177996
------------------------------------------------------------------------------
/* alternative method using xi3/
NOTE: You will need to download xi3, which can be installed by typing findit xi3 in the command line (see How can I use the findit command to search for programs and get additional help? for more information about using findit).
xi3: stcox e.agegrp, nohr
d.agegrp _Iagegrp_1-4 (naturally coded; _Iagegrp_1 omitted)
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(3) = 19.56
Log likelihood = -289.41333 Prob > chi2 = 0.0002
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
_Iagegrp_2 | .1042989 .1920546 0.54 0.587 -.2721212 .480719
_Iagegrp_3 | .2206278 .2058906 1.07 0.284 -.1829103 .624166
_Iagegrp_4 | .7677369 .2093199 3.67 0.000 .3574775 1.177996
------------------------------------------------------------------------------
Table 4.8, page 129.
stcox age, nohr
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(1) = 21.35
Log likelihood = -288.51804 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age | .0814042 .0174352 4.67 0.000 .0472317 .1155766
------------------------------------------------------------------------------
Table 4.9 1.4, page 133.
use uis, clear
stset time, failure(censor)
generate drug = ivhx
recode drug 1=0 2/3=1
stcox drug if age~=., nohr
Cox regression -- Breslow method for ties
No. of subjects = 605 Number of obs = 605
No. of failures = 489
Time at risk = 144822
LR chi2(1) = 11.78
Log likelihood = -2825.9664 Prob > chi2 = 0.0006
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
drug | .3210045 .0948107 3.39 0.001 .1351789 .5068301
------------------------------------------------------------------------------
stcox drug age, nohr
Cox regression -- Breslow method for ties
No. of subjects = 605 Number of obs = 605
No. of failures = 489
Time at risk = 144822
LR chi2(2) = 23.32
Log likelihood = -2820.1989 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
drug | .4394466 .1007151 4.36 0.000 .2420487 .6368445
age | -.0263792 .0078392 -3.37 0.001 -.0417438 -.0110147
------------------------------------------------------------------------------
Table 4.10, page 135.
stcox treat if age~=., nohr
Cox regression -- Breslow method for ties
No. of subjects = 623 Number of obs = 623
No. of failures = 504
Time at risk = 146816
LR chi2(1) = 6.05
Log likelihood = -2930.087 Prob > chi2 = 0.0139
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
treat | -.2196282 .0893249 -2.46 0.014 -.3947018 -.0445546
------------------------------------------------------------------------------
stcox treat age, nohr
Cox regression -- Breslow method for ties
No. of subjects = 623 Number of obs = 623
No. of failures = 504
Time at risk = 146816
LR chi2(2) = 9.48
Log likelihood = -2928.3748 Prob > chi2 = 0.0088
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
treat | -.2229808 .0893303 -2.50 0.013 -.3980651 -.0478966
age | -.013268 .0072117 -1.84 0.066 -.0274027 .0008667
------------------------------------------------------------------------------
generate trtxage = treat*age
stcox treat age trtxage, nohr
Cox regression -- Breslow method for ties
No. of subjects = 623 Number of obs = 623
No. of failures = 504
Time at risk = 146816
LR chi2(3) = 12.04
Log likelihood = -2927.0915 Prob > chi2 = 0.0072
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
treat | .5227169 .4744893 1.10 0.271 -.407265 1.452699
age | -.0017705 .0101253 -0.17 0.861 -.0216156 .0180747
trtxage | -.0231941 .0145004 -1.60 0.110 -.0516144 .0052261
------------------------------------------------------------------------------
Figure 4.2, page 139.
use hmohiv, clear
stset time, failure(censor)
stcox drug, nohr bases(bs0)
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(1) = 10.20
Log likelihood = -294.09639 Prob > chi2 = 0.0014
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
drug | .7792305 .2422582 3.22 0.001 .3044132 1.254048
------------------------------------------------------------------------------
generate bs1 = bs0^exp(.7792305)
graph twoway scatter bs0 bs1 time, connect(J J) sort ylabel(0 1) xlabel(0(10)60)
Figure 4.3, page 141.
replace bs0=. if drug==1 replace bs1=. if drug==0 graph twoway scatter bs0 bs1 time, connect(J J) sort ylabel(0 1) xlabel(0(10)60)
Table 4.12, page 143.
generate age35 = age-35
stcox drug age35, nohr bases(bs0_35)
Cox regression -- Breslow method for ties
No. of subjects = 100 Number of obs = 100
No. of failures = 80
Time at risk = 1136
LR chi2(2) = 34.98
Log likelihood = -281.70404 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
drug | .9413856 .2555104 3.68 0.000 .4405943 1.442177
age35 | .0915319 .0184879 4.95 0.000 .0552963 .1277675
------------------------------------------------------------------------------
Fig. 4.4, page 144.
gen bs1_35 = bs0_35^exp(.9413856) gen bs0_35a=bs0_35 if drug==0 gen bs1_35a=bs1_35 if drug==1 label variable bs0_35a "IV Drug Use Present" label variable bs1_35a "IV Drug Use Absent" graph twoway scatter bs0_35a bs1_35a time, connect(J J) sort ylabel(0(.25)1) xlabel(0(10)60)
Fig. 4.5a, page 146.
gen age30 = age - 30 quietly quietly stcox drug age30, nohr bases(bs0_30) gen bs1_30 = bs0_30^exp(.9413856 ) label variable bs0_30 "IV Drug Use Present" label variable bs1_30 "IV Drug Use Absent" graph twoway scatter bs0_30 bs1_30 time, connect(J J) sort ylabel(0 1) xlabel(0(10)60)
Fig. 4.5b, page 146.
Note: We created the variables used in the graph in the code for table 4.12 and fig. 4.4.
label variable bs0_35 "IV Drug Use Present" label variable bs1_35 "IV Drug Use Absent" graph twoway scatter bs0_35 bs1_35 time, connect(J J) sort ylabel(0(.25)1) xlabel(0(10)60)
Fig. 4.5c, page 146.
gen age40 = age - 40 quietly stcox drug age40, nohr bases(bs0_40) gen bs1_40 = bs0_40^exp(.9413856 ) label variable bs0_40 "IV Drug Use Present" label variable bs1_40 "IV Drug Use Absent" graph twoway scatter bs0_40 bs1_40 time, connect(J J) sort ylabel(0(.25)1) xlabel(0(10)60)
Fig. 4.5d, page 146.
gen age45 = age - 45 quietly stcox drug age45, nohr bases(bs0_45) gen bs1_45 = bs0_45^exp(.9413856 ) label variable bs0_45 "IV Drug Use Present" label variable bs1_45 "IV Drug Use Absent" graph twoway scatter bs0_45 bs1_45 time, connect(J J) sort ylabel(0(.25)1) xlabel(0(10)60)
Table 4.13 and fig. 4.6, page 148.
use uis
stset time, failure(censor)
generate agec = age-30
generate ndrgtxc = ndrugtx-3
generate drug = ivhx
recode drug 1=0 2/3=1
stcox treat agec drug ndrgtxc, nohr bases(bs0)
Cox regression -- Breslow method for ties
No. of subjects = 593 Number of obs = 593
No. of failures = 481
Time at risk = 141069
LR chi2(4) = 41.09
Log likelihood = -2753.4247 Prob > chi2 = 0.0000
------------------------------------------------------------------------------
_t |
_d | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
treat | -.2271044 .0915798 -2.48 0.013 -.4065975 -.0476113
agec | -.0307434 .0079358 -3.87 0.000 -.0462974 -.0151895
drug | .3425819 .1042631 3.29 0.001 .1382299 .5469339
ndrgtxc | .0309095 .0079891 3.87 0.000 .0152511 .0465679
------------------------------------------------------------------------------
gen bs1 = bs0^exp( -.2271044 )
label variable bs0 "Treat = 0"
label variable bs1 "Treat = 1"
graph twoway scatter bs0 bs1 time, msymbol(i i) connect(l l) sort ylabel(0(.25)1) xlabel(0(500)1500)
Fig. 4.8, page 155. We currently don't know how to do this example.
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
