--------------------------------------------------------------------------------------------------
name: <unnamed>
log: C:\Users\mexmi\Documents\newer web pages\Soc_382\logs\event_history_log1.log
log type: text
opened on: 19 Feb 2019, 10:20:26
. use "C:\Users\mexmi\Documents\current class files\intro soc methods\Soc 382 stuff\Soc 382 event history\HCMST file for class, single observation per couple v2.dta", clear
( )
* This is a brief log for examining the single observation event history dataset. More commentary in the Excel file.
. tabulate ever_broke_up_w2345
whether |
subject who |
was |
followed up |
ever broke |
up or not | Freq. Percent Cum.
------------+-----------------------------------
0 | 2,167 81.16 81.16
1 | 503 18.84 100.00
------------+-----------------------------------
Total | 2,670 100.00
* 503 breakups out of 2670 total couples.
*Now I issue the stset command, which calls up the stset command that has already been applied to this dataset.
. stset
-> stset relationship_duration_at_end, failure(ever_broke_up_w2345==1)
enter(time how_long_relationship_w1)
failure event: ever_broke_up_w2345 == 1
obs. time interval: (0, relationship_duration_at_end]
enter on or after: time how_long_relationship_w1
exit on or before: failure
------------------------------------------------------------------------------
2670 total observations
33 event time missing (relationship_duration_at_end>=.) PROBABLE ERROR
------------------------------------------------------------------------------
2637 observations remaining, representing
494 failures in single-record/single-failure data
8265.833 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 0
last observed exit t = 73.33334
*Now a Kaplan-Meier graph, for survival of couples as intact couples, by whether they met online or not.
. sts graph if met_on_after_95==1 & same_sex_couple_v2==0 & _t<20, by(met_online_augmented)
failure _d: ever_broke_up_w2345 == 1
analysis time _t: relationship_duration_at_end
enter on or after: time how_long_relationship_w1
* Now an stcox proportional hazards model predicting breakup. Note that there is no dependent variable (because the stset command already told Stata what the outcome of interest is) and there is no control for the time axis, relationship duration, because the Cox proportional hazard model controls for that automatically. Also note that as this is the single observation dataset, all covariates are fixed at time 1.
. stcox resp_col_dgre married coresident children_in_hh met_online_augmented ln_hhinc_2009dollars
failure _d: ever_broke_up_w2345 == 1
analysis time _t: relationship_duration_at_end
enter on or after: time how_long_relationship_w1
Iteration 0: log likelihood = -2535.3568
Iteration 1: log likelihood = -2423.3912
Iteration 2: log likelihood = -2421.8666
Iteration 3: log likelihood = -2421.8658
Refining estimates:
Iteration 0: log likelihood = -2421.8658
Cox regression -- Breslow method for ties
No. of subjects = 2582 Number of obs = 2582
No. of failures = 478
Time at risk = 8121.833358
LR chi2(6) = 226.98
Log likelihood = -2421.8658 Prob > chi2 = 0.0000
--------------------------------------------------------------------------------------
_t | Haz. Ratio Std. Err. z P>|z| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
resp_col_dgre | .8267504 .0831387 -1.89 0.059 .6788549 1.006867
married | .3542358 .0529819 -6.94 0.000 .2642293 .474902
coresident | .3391017 .0436468 -8.40 0.000 .263493 .4364059
children_in_hh | 1.112869 .0609199 1.95 0.051 .9996501 1.23891
met_online_augmented | 1.095103 .131106 0.76 0.448 .8660613 1.384717
ln_hhinc_2009dollars | .9061744 .0515513 -1.73 0.083 .810565 1.013061
--------------------------------------------------------------------------------------
. regress relationship_duration_at_end resp_col_dgre married coresident children_in_hh met_online_
> augmented ln_hhinc_2009dollars if ever_broke_up_w2345==1
Source | SS df MS Number of obs = 478
-------------+------------------------------ F( 6, 471) = 20.55
Model | 8558.76186 6 1426.46031 Prob > F = 0.0000
Residual | 32697.7652 471 69.4220068 R-squared = 0.2075
-------------+------------------------------ Adj R-squared = 0.1974
Total | 41256.527 477 86.4916709 Root MSE = 8.332
--------------------------------------------------------------------------------------
relationship_durat~d | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
resp_col_dgre | 1.187367 .8458316 1.40 0.161 -.474703 2.849438
married | 8.886208 1.058656 8.39 0.000 6.805935 10.96648
coresident | .1817787 .8803307 0.21 0.836 -1.548083 1.91164
children_in_hh | -.9104321 .4433885 -2.05 0.041 -1.781696 -.0391679
met_online_augmented | -4.684189 .9610017 -4.87 0.000 -6.57257 -2.795807
ln_hhinc_2009dollars | -.9450353 .4867864 -1.94 0.053 -1.901577 .0115065
_cons | 17.82119 5.167315 3.45 0.001 7.667347 27.97503
--------------------------------------------------------------------------------------
. *The regression on failure time for couples who broke up is bad in EVERY way. The badness of this is one reason we do Event History Analysis.
. clear all
. use "C:\Users\mexmi\Documents\current class files\intro soc methods\Soc 382 stuff\Soc 382 event
> history\wave1 to 5 combined for sharing with class stset v3 reduced vars.dta", clear
( )
. log close
name: <unnamed>
log: C:\Users\mexmi\Documents\newer web pages\Soc_382\logs\event_history_log1.log
log type: text
closed on: 19 Feb 2019, 11:56:39
--------------------------------------------------------------------------------------------------