|
|
|
||||
|
|
|||||
Table 13.1, page 471.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
sort seconds
list seconds censor
seconds censor
1. 1.41 1
2. 1.41 0
3. 1.51 0
4. 1.67 0
5. 1.68 0
6. 1.86 0
7. 2.12 0
.. output omitted ...
50. 6.21 1
51. 6.3 0
52. 6.6 1
53. 7.2 0
54. 9.59 0
55. 12.29 0
56. 13.18 0
57. 17.15 1
Table 13.2, page 477.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
generate event = ~censor
ltable seconds event, noadjust intervals(0 1 2 3 4 5 6 7 8 18)
Beg. Std.
Interval Total Deaths Lost Survival Error [95% Conf. Int.]
-------------------------------------------------------------------------------
1 2 57 5 1 0.9123 0.0375 0.8020 0.9625
2 3 51 14 3 0.6619 0.0632 0.5224 0.7693
3 4 34 9 2 0.4867 0.0683 0.3481 0.6117
4 5 23 6 4 0.3597 0.0673 0.2314 0.4896
5 6 13 2 2 0.3044 0.0674 0.1803 0.4378
6 7 9 2 2 0.2367 0.0673 0.1199 0.3757
7 8 5 1 0 0.1894 0.0685 0.0783 0.3373
8 18 4 3 1 0.0473 0.0444 0.0038 0.1884
-------------------------------------------------------------------------------
ltable seconds event, noadjust hazard intervals(0 1 2 3 4 5 6 7 8 18)
Beg. Cum. Std. Std.
Interval Total Failure Error Hazard Error [95% Conf. Int.]
-------------------------------------------------------------------------------
1 2 57 0.0877 0.0375 0.0877 0.0392 0.0285 0.1797
2 3 51 0.3381 0.0632 0.2745 0.0734 0.1501 0.4359
3 4 34 0.5133 0.0683 0.2647 0.0882 0.1210 0.4636
4 5 23 0.6403 0.0673 0.2609 0.1065 0.0957 0.5073
5 6 13 0.6956 0.0674 0.1538 0.1088 0.0186 0.4286
6 7 9 0.7633 0.0673 0.2222 0.1571 0.0269 0.6191
7 8 5 0.8106 0.0685 0.2000 0.2000 0.0051 0.7378
8 18 4 0.9527 0.0444 0.0750 0.0433 0.0155 0.1806
-------------------------------------------------------------------------------
ltable seconds event, intervals(0 1 2 3 4 5 6 7 8 18)
Beg. Std.
Interval Total Deaths Lost Survival Error [95% Conf. Int.]
-------------------------------------------------------------------------------
1 2 57 5 1 0.9115 0.0378 0.8004 0.9622
2 3 51 14 3 0.6537 0.0643 0.5121 0.7633
3 4 34 9 2 0.4754 0.0690 0.3362 0.6021
4 5 23 6 4 0.3396 0.0680 0.2116 0.4719
5 6 13 2 2 0.2830 0.0674 0.1608 0.4181
6 7 9 2 2 0.2122 0.0666 0.0998 0.3526
7 8 5 1 0 0.1698 0.0654 0.0662 0.3140
8 18 4 3 1 0.0243 0.0331 0.0005 0.1634
-------------------------------------------------------------------------------
ltable seconds event, hazard intervals(0 1 2 3 4 5 6 7 8 18)
Beg. Cum. Std. Std.
Interval Total Failure Error Hazard Error [95% Conf. Int.]
-------------------------------------------------------------------------------
1 2 57 0.0885 0.0378 0.0926 0.0414 0.0115 0.1737
2 3 51 0.3463 0.0643 0.3294 0.0868 0.1592 0.4996
3 4 34 0.5246 0.0690 0.3158 0.1039 0.1121 0.5195
4 5 23 0.6604 0.0680 0.3333 0.1342 0.0703 0.5963
5 6 13 0.7170 0.0674 0.1818 0.1280 0.0000 0.4328
6 7 9 0.7878 0.0666 0.2857 0.2000 0.0000 0.6776
7 8 5 0.8302 0.0654 0.2222 0.2208 0.0000 0.6551
8 18 4 0.9757 0.0331 0.1500 0.0573 0.0377 0.2623
-------------------------------------------------------------------------------
Figure 13.1, page 479.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
gen event = ~censor
ltable seconds event, noadjust intervals(0 1 2 3 4 5 6 7 8 18) saving(ltable1)
use ltable1, clear
set obs 9
replace t1 = 1 if _n==9
replace survival = 1 if _n==9
sort t1
twoway (line survival t1), xtitle("Seconds after light turns green") ytitle("Estimated S(t)") xlabel(0(5)20)
twoway (line hazard t1), xtitle("Seconds after light turns green") ytitle("Estimated h(t)") xlabel(0(5)20)
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
gen event = ~censor
ltable seconds event, intervals(0 1 2 3 4 5 6 7 8 18) saving(ltable2)
use ltable2, clear
set obs 9
replace t1 = 1 if _n==9
replace survival = 1 if _n==9
sort t1
twoway scatter survival t1, connect(stepstair) ///
xtitle("Seconds after light turns green") ytitle("Estimated S(t)")
twoway scatter hazard t1, connect(stepstair) ///
xtitle("Seconds after light turns green") ytitle("Estimated h(t)")
Table 13.3, page 484.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
generate event = ~censor
stset seconds, failure(event)
sts list
failure _d: event
analysis time _t: seconds
Beg. Net Survivor Std.
Time Total Fail Lost Function Error [95% Conf. Int.]
-------------------------------------------------------------------------------
1.41 57 1 1 0.9825 0.0174 0.8819 0.9975
1.51 55 1 0 0.9646 0.0246 0.8657 0.9910
1.67 54 1 0 0.9467 0.0299 0.8439 0.9825
1.68 53 1 0 0.9289 0.0343 0.8215 0.9727
1.86 52 1 0 0.9110 0.0380 0.7993 0.9620
2.12 51 1 0 0.8931 0.0412 0.7775 0.9505
2.19 50 1 0 0.8753 0.0441 0.7561 0.9385
2.36 49 0 1 0.8753 0.0441 0.7561 0.9385
2.48 48 1 0 0.8570 0.0468 0.7344 0.9258
2.5 47 1 0 0.8388 0.0492 0.7130 0.9127
2.53 46 1 0 0.8206 0.0514 0.6921 0.8992
2.54 45 1 0 0.8023 0.0534 0.6715 0.8854
2.56 44 1 0 0.7841 0.0552 0.6511 0.8712
2.62 43 1 0 0.7659 0.0569 0.6311 0.8568
2.68 42 1 0 0.7476 0.0584 0.6113 0.8421
2.76 41 0 1 0.7476 0.0584 0.6113 0.8421
2.78 40 0 1 0.7476 0.0584 0.6113 0.8421
2.83 39 1 0 0.7285 0.0599 0.5903 0.8266
2.88 38 1 0 0.7093 0.0614 0.5697 0.8108
2.89 37 1 0 0.6901 0.0626 0.5493 0.7948
2.92 36 1 0 0.6710 0.0637 0.5292 0.7786
2.98 35 1 0 0.6518 0.0647 0.5094 0.7621
...
3.57 26 1 0 0.5121 0.0692 0.3701 0.6372
3.58 25 1 0 0.4916 0.0694 0.3506 0.6181
...
4.96 14 1 0 0.3349 0.0683 0.2067 0.4681
5.12 13 0 1 0.3349 0.0683 0.2067 0.4681
5.39 12 1 0 0.3070 0.0681 0.1815 0.4416
5.73 11 1 0 0.2791 0.0674 0.1574 0.4144
5.88 10 0 1 0.2791 0.0674 0.1574 0.4144
6.03 9 1 0 0.2481 0.0666 0.1308 0.3846
6.21 8 0 1 0.2481 0.0666 0.1308 0.3846
6.3 7 1 0 0.2126 0.0659 0.1011 0.3514
6.6 6 0 1 0.2126 0.0659 0.1011 0.3514
7.2 5 1 0 0.1701 0.0650 0.0670 0.3133
9.59 4 1 0 0.1276 0.0611 0.0388 0.2711
12.29 3 1 0 0.0851 0.0535 0.0172 0.2244
13.18 2 1 0 0.0425 0.0403 0.0034 0.1730
17.15 1 0 1 0.0425 0.0403 0.0034 0.1730
-------------------------------------------------------------------------------
Figure 13.2, page 485. We are only able to generate the top plot for now.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear gen event = ~censor stset seconds, failure(event) sts graph
Figure 13.4, page 493.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
gen event = ~censor
stset seconds, failure(event)
sts generate surv = s nelson = na
gen nlsurv = -log(surv)
sort seconds
twoway (scatter nelson seconds, connect(stepstair) lpattern(solid) msize(zero)) ///
(scatter nlsurv seconds, connect(stepstair) lpattern(dash) msize(zero)), ///
xtitle("Seconds after light turns green") ytitle("Estimated H(t)") ///
legend(ring(0) pos(5) lab(1 "Nelson-Aalen") lab(2 "Negative Log Survival"))
Figure 13.5, page 496. Our graphs do not exactly match those in the book, most likely due to a difference in kernel functions.
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/honking, clear
gen event = ~censor
stset seconds, failure(event)
sts graph, haz width(.5) ylabel(0(.1).4) xtitle("Seconds after light turns green")
sts graph, haz width(1) ylabel(0(.1).4) xtitle("Seconds after light turns green")
sts graph, haz width(1.5) ylabel(0(.1).4) xtitle("Seconds after light turns green")
Figure 13.6, page 499. Again, due to a difference in kernel functions, our smoothed hazard graphs do not exactly match those in the text.
/* Column 1 */
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/alcohol_relapse, clear
gen event = ~censor
stset weeks, failure(event)
sts graph, xtitle("Weeks since discharge") legend(ring(0) pos(2)) xlabel(0(25)100) ylabel(,nogrid) scheme(Lean1)
sts graph, cumh xtitle("Weeks since discharge") legend(ring(0) pos(5)) xlabel(0(25)100) ylabel(,nogrid) scheme(Lean1)
sts graph, haz width(6) xtitle("Weeks since discharge") legend(ring(0) pos(5)) xlabel(0(25)100) ylabel(,nogrid) scheme(Lean1)
/* Column 2 */
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/judges, clear
stset tenure, failure(leave)
sts graph, xtitle("Years on Court") legend(ring(0) pos(2)) xlabel(0(5)35) ylabel(,nogrid) scheme(Lean1)
sts graph, cumh xtitle("Years on Court") legend(ring(0) pos(5)) xlabel(0(5)35) ylabel(,nogrid) scheme(Lean1)
sts graph, haz width(5) xtitle("Years on Court") legend(ring(0) pos(5)) xlabel(0(5)35) ylabel(0(.02).14,nogrid) scheme(Lean1)
/* Column 3 */
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/firstdepression, clear
gen event = ~censor
stset age, failure(event)
sts graph, xtitle("Age (in years)") legend(ring(0) pos(2)) xlabel(0(25)100) ylabel(,nogrid) scheme(Lean1)
sts graph, cumh xtitle("Age (in years)") legend(ring(0) pos(5)) xlabel(0(25)100) ylabel(,nogrid) scheme(Lean1)
sts graph, haz width(3.5) xtitle("Age (in years)") legend(ring(0) pos(5)) xlabel(0(25)100) ylabel(0(.001).003,nogrid) scheme(Lean1)
/* Column 4 */
use http://www.ats.ucla.edu/stat/stata/examples/alda/data/healthworkers, clear
gen event = ~censor
stset weeks, failure(event)
sts graph, xtitle("Weeks since hired") legend(ring(0) pos(2)) xlabel(0(26)130) ylabel(,nogrid) scheme(Lean1)
sts graph, cumh xtitle("Age (in years)") legend(ring(0) pos(5)) xlabel(0(26)130) ylabel(,nogrid) scheme(Lean1)
sts graph, haz width(6) xtitle("Age (in years)") legend(ring(0) pos(5)) xlabel(0(26)130) ylabel(0(.005).01,nogrid) scheme(Lean1)
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