DEM 7223 Homework 1
Steven Weeks
10/15/2019
## Installing package into '/storage01/users/izf381/R/x86_64-redhat-linux-gnu-library/3.6'
## (as 'lib' is unspecified)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union## Loading required package: carData##
## Attaching package: 'car'## The following object is masked from 'package:dplyr':
##
## recode##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numericDefinitions
1) Define your event variable - The event variable is the transition of employement in Wave 3 to unemployment (fired, layed off, let go) in Wave 4
2) Define a duration or time variable - If the age at unemployment is censored, then the age at the date of interview is the censored age at unemployment, or if not censored, the actual age at umeployment.
3) Define a censoring indicator - If someone has experienced the event the are assigned a “0”, if they experience an event they are assigned a “1”.
Descriptive estimates
## Loading required package: ggplot2## Registered S3 methods overwritten by 'ggplot2':
## method from
## [.quosures rlang
## c.quosures rlang
## print.quosures rlang## Loading required package: ggpubr## Loading required package: magrittr4) Estimate the following functions of survival time for that variable and plot them:
a. Survival
## Call: survfit(formula = Surv(time = job.loss.age, event = job_transition) ~
## 1, data = addhealth)
##
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 7101 114 0.984 0.00149 0.981 0.987
## 20 6987 338 0.936 0.00290 0.931 0.942
## 21 6649 344 0.888 0.00374 0.881 0.895
## 22 6305 469 0.822 0.00454 0.813 0.831
## 23 5836 400 0.766 0.00503 0.756 0.775
## 24 5436 333 0.719 0.00534 0.708 0.729
## 25 5102 84 0.707 0.00540 0.696 0.717
## 26 4981 27 0.703 0.00542 0.692 0.714
b. Hazard
## $time
## [1] 19.5 20.5 21.5 22.5 23.5 24.5 25.5
##
## $haz
## [1] 0.049584812 0.053123498 0.077297339 0.071002134 0.063216305 0.016656801
## [7] 0.005818767
##
## $var
## [1] 7.275614e-06 8.205726e-06 1.274596e-05 1.260855e-05 1.200492e-05
## [6] 3.303130e-06 1.256180e-06
## time haz var
## 1 19.5 0.049584812 7.275614e-06
## 2 20.5 0.053123498 8.205726e-06
## 3 21.5 0.077297339 1.274596e-05
## 4 22.5 0.071002134 1.260855e-05
## 5 23.5 0.063216305 1.200492e-05
## 6 24.5 0.016656801 3.303130e-06
## 7 25.5 0.005818767 1.256180e-06c. Cumulative Hazard
### 5) Carry out the following analysis #### a. Kaplan-Meier survival analysis of the outcome
## Call: survfit(formula = Surv(job.loss.age, job_transition) ~ 1, data = addhealth)
##
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 7101 114 0.984 0.00149 0.981 0.987
## 20 6987 338 0.936 0.00290 0.931 0.942
## 21 6649 344 0.888 0.00374 0.881 0.895
## 22 6305 469 0.822 0.00454 0.813 0.831
## 23 5836 400 0.766 0.00503 0.756 0.775
## 24 5436 333 0.719 0.00534 0.708 0.729
## 25 5102 84 0.707 0.00540 0.696 0.717
## 26 4981 27 0.703 0.00542 0.692 0.714b. Define a grouping variable, this can be dichotomous or categorical.
The grouping variabe is sexual orientation
d. Comparison of Kaplan-Meier survival across grouping variable in your data. Interpret your results. It would appear across all groups that the age at highest risk is around 22-23 years old.
Job transition based on sexual orienation
## Call: survfit(formula = Surv(job.loss.age, job_transition) ~ sexorient,
## data = addhealth)
##
## sexorient=a_straight
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 6866 111 0.984 0.00152 0.981 0.987
## 20 6755 320 0.937 0.00293 0.932 0.943
## 21 6435 330 0.889 0.00379 0.882 0.897
## 22 6105 450 0.824 0.00460 0.815 0.833
## 23 5655 386 0.767 0.00510 0.757 0.777
## 24 5269 318 0.721 0.00541 0.711 0.732
## 25 4950 79 0.710 0.00548 0.699 0.720
## 26 4835 26 0.706 0.00550 0.695 0.717
##
## sexorient=b_bisexual
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 122 3 0.975 0.0140 0.948 1.000
## 20 119 9 0.902 0.0270 0.850 0.956
## 21 110 10 0.820 0.0348 0.754 0.891
## 22 100 15 0.697 0.0416 0.620 0.783
## 23 85 2 0.680 0.0422 0.602 0.768
## 24 83 6 0.631 0.0437 0.551 0.723
## 25 77 3 0.607 0.0442 0.526 0.700
## 26 74 1 0.598 0.0444 0.517 0.692
##
## sexorient=c_LGB
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 20 113 9 0.920 0.0255 0.872 0.972
## 21 104 4 0.885 0.0300 0.828 0.946
## 22 100 4 0.850 0.0336 0.786 0.918
## 23 96 12 0.743 0.0411 0.667 0.828
## 24 84 9 0.664 0.0444 0.582 0.757
## 25 75 2 0.646 0.0450 0.564 0.740
## observed expected o-e
## a_straight 2020 2042.75755 -22.757546
## b_bisexual 49 33.01654 15.983458
## c_LGB 40 33.22591 6.774089## Call: survfit(formula = Surv(time = job.loss.age, event = job_transition) ~
## sexorient + sex, data = addhealth)
##
## sexorient=a_straight, sex=a_male
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 3245 63 0.981 0.00242 0.976 0.985
## 20 3182 160 0.931 0.00444 0.923 0.940
## 21 3022 173 0.878 0.00575 0.867 0.889
## 22 2849 251 0.801 0.00701 0.787 0.814
## 23 2598 213 0.735 0.00775 0.720 0.750
## 24 2385 189 0.677 0.00821 0.661 0.693
## 25 2196 47 0.662 0.00830 0.646 0.679
## 26 2140 21 0.656 0.00834 0.640 0.672
##
## sexorient=a_straight, sex=b_female
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 3621 48 0.987 0.00190 0.983 0.990
## 20 3573 160 0.943 0.00387 0.935 0.950
## 21 3413 157 0.899 0.00500 0.889 0.909
## 22 3256 199 0.844 0.00603 0.833 0.856
## 23 3057 173 0.796 0.00669 0.783 0.810
## 24 2884 129 0.761 0.00709 0.747 0.775
## 25 2754 32 0.752 0.00718 0.738 0.766
## 26 2695 5 0.751 0.00719 0.737 0.765
##
## sexorient=b_bisexual, sex=a_male
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 19 1 0.947 0.0512 0.852 1.000
## 20 18 1 0.895 0.0704 0.767 1.000
## 21 17 1 0.842 0.0837 0.693 1.000
## 24 16 2 0.737 0.1010 0.563 0.964
##
## sexorient=b_bisexual, sex=b_female
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 19 103 2 0.981 0.0136 0.954 1.000
## 20 101 8 0.903 0.0292 0.848 0.962
## 21 93 9 0.816 0.0382 0.744 0.894
## 22 84 15 0.670 0.0463 0.585 0.767
## 23 69 2 0.650 0.0470 0.565 0.749
## 24 67 4 0.612 0.0480 0.524 0.713
## 25 63 3 0.583 0.0486 0.495 0.686
## 26 60 1 0.573 0.0487 0.485 0.677
##
## sexorient=c_LGB, sex=a_male
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 20 71 4 0.944 0.0274 0.892 0.999
## 21 67 1 0.930 0.0304 0.872 0.991
## 22 66 4 0.873 0.0395 0.799 0.954
## 23 62 8 0.761 0.0506 0.668 0.867
## 24 54 6 0.676 0.0555 0.576 0.794
## 25 48 2 0.648 0.0567 0.546 0.769
##
## sexorient=c_LGB, sex=b_female
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 20 42 5 0.881 0.0500 0.788 0.985
## 21 37 3 0.810 0.0606 0.699 0.937
## 23 34 4 0.714 0.0697 0.590 0.865
## 24 30 3 0.643 0.0739 0.513 0.805e. Plot the hazard function for the analysis for each level of the group variable (continued assignment from part d)
#Part B #Parametric models ### 1) Carry out the following analysis: Define your outcome as in part A. Also consider what covariates are hypothesized to affect the outcome variable. Define these and construct a parametric model for your outcome.
####The outcome variable is the transition of employement in Wave 3 to unemployment (fired, layed off, let go) in Wave 4 ####The predictior variables are chosen based on previous literature and what may effect lossing a job for LGB individuals. The variables are education, depression, insurance status, getting beaten up, alcohol consumption, general health, race/ethnicity, sex, how a job was lost, sexual orientation, unable to pay utilities, and job satisfaction
Fit the parametric model of your choosing to the data.
a. Did you choose an AFT or PH model and why?
I chose a Cox Proportional Hazard Model to analyzing the effect of several covariates after looking at the results of exponential, Weibull, and piecewise exponential models. The Weibull had the lowest AIC score and fit somewhat, but the Cox model proved to be better.
b. Justify what parametric distribution you choose
c. Carry out model fit diagnostics for the model
d. Include all main effects in the model
e. Test for an interaction between at least two of the predictors
f. Interpret your results and write them up
## Loading required package: grid## Loading required package: Matrix##
## Attaching package: 'survey'## The following object is masked from 'package:graphics':
##
## dotchart## Stratified 1 - level Cluster Sampling design (with replacement)
## With (130) clusters.
## svydesign(ids = ~psuscid, strata = ~region, weights = ~GSWGT4,
## data = addhealth, nest = T)## Call:
## svycoxph(formula = Surv(time = job.loss.age, event = job_transition) ~
## sex + sexorient, design = des2)
##
## n= 7101, number of events= 2109
##
## coef exp(coef) se(coef) z Pr(>|z|)
## sexb_female -0.40943 0.66403 0.05377 -7.615 2.64e-14 ***
## sexorientb_bisexual 0.55580 1.74333 0.17214 3.229 0.00124 **
## sexorientc_LGB 0.10467 1.11034 0.21069 0.497 0.61934
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## exp(coef) exp(-coef) lower .95 upper .95
## sexb_female 0.664 1.5060 0.5976 0.7378
## sexorientb_bisexual 1.743 0.5736 1.2441 2.4429
## sexorientc_LGB 1.110 0.9006 0.7347 1.6780
##
## Concordance= 0.555 (se = 0.007 )
## Likelihood ratio test= NA on 3 df, p=NA
## Wald test = 69.8 on 3 df, p=5e-15
## Score (logrank) test = NA on 3 df, p=NA
###Cox model with outcome and predictor variables.
## Stratified 1 - level Cluster Sampling design (with replacement)
## With (130) clusters.
## svydesign(ids = ~psuscid, strata = ~region, weights = ~GSWGT4,
## data = addhealth, nest = T)## Call:
## svycoxph(formula = Surv(time = job.loss.age, event = job_transition) ~
## sex + racethnic + sexorient + educ + unable_pay_utilities +
## insurance_statusW4 + general_health + depressionW4 +
## alcohol_day_permonthW4 + leave_job + job_satisfaction +
## beaten_upW4, design = des2)
##
## n= 7101, number of events= 2109
##
## coef exp(coef) se(coef) z
## sexb_female -0.335328 0.715103 0.060652 -5.529
## racethnicb-nhblack 0.252315 1.287002 0.088078 2.865
## racethnicc-hispanic -0.001349 0.998652 0.120087 -0.011
## racethnicd-asian -0.281351 0.754763 0.143848 -1.956
## racethnice-native_american 0.392680 1.480944 0.333665 1.177
## racethnicf-other 0.356611 1.428481 0.332484 1.073
## sexorientb_bisexual 0.324764 1.383704 0.175884 1.846
## sexorientc_LGB 0.149616 1.161388 0.200128 0.748
## educb_highschool_grad -0.135556 0.873230 0.102126 -1.327
## educc_college_bach -0.604396 0.546404 0.147919 -4.086
## educd_college+ -1.165803 0.311672 0.157958 -7.380
## unable_pay_utilitiesb_yes 0.322562 1.380661 0.068373 4.718
## insurance_statusW4b_yes_insurance -0.425704 0.653309 0.062529 -6.808
## general_healthb_poor/bad 0.150772 1.162731 0.097135 1.552
## depressionW4b_yes 0.174849 1.191066 0.082055 2.131
## alcohol_day_permonthW4b_1or2days/week -0.003821 0.996186 0.072122 -0.053
## alcohol_day_permonthW4c_3to5days/week -0.005482 0.994533 0.094637 -0.058
## alcohol_day_permonthW4d_daily 0.235969 1.266135 0.137049 1.722
## leave_jobb_health -1.179450 0.307448 0.218042 -5.409
## leave_jobc_other -0.928580 0.395114 0.135977 -6.829
## leave_jobd_skip -0.805064 0.447059 0.103178 -7.803
## job_satisfactionb_neutral 0.111726 1.118206 0.080248 1.392
## job_satisfactionc_dissatisfied 0.199357 1.220618 0.094520 2.109
## beaten_upW4b_yes 0.019293 1.019480 0.099590 0.194
## Pr(>|z|)
## sexb_female 3.22e-08 ***
## racethnicb-nhblack 0.00417 **
## racethnicc-hispanic 0.99103
## racethnicd-asian 0.05048 .
## racethnice-native_american 0.23925
## racethnicf-other 0.28346
## sexorientb_bisexual 0.06483 .
## sexorientc_LGB 0.45470
## educb_highschool_grad 0.18440
## educc_college_bach 4.39e-05 ***
## educd_college+ 1.58e-13 ***
## unable_pay_utilitiesb_yes 2.39e-06 ***
## insurance_statusW4b_yes_insurance 9.89e-12 ***
## general_healthb_poor/bad 0.12062
## depressionW4b_yes 0.03310 *
## alcohol_day_permonthW4b_1or2days/week 0.95775
## alcohol_day_permonthW4c_3to5days/week 0.95380
## alcohol_day_permonthW4d_daily 0.08511 .
## leave_jobb_health 6.33e-08 ***
## leave_jobc_other 8.55e-12 ***
## leave_jobd_skip 6.06e-15 ***
## job_satisfactionb_neutral 0.16384
## job_satisfactionc_dissatisfied 0.03493 *
## beaten_upW4b_yes 0.84639
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## exp(coef) exp(-coef) lower .95
## sexb_female 0.7151 1.3984 0.6350
## racethnicb-nhblack 1.2870 0.7770 1.0829
## racethnicc-hispanic 0.9987 1.0014 0.7892
## racethnicd-asian 0.7548 1.3249 0.5693
## racethnice-native_american 1.4809 0.6752 0.7701
## racethnicf-other 1.4285 0.7000 0.7445
## sexorientb_bisexual 1.3837 0.7227 0.9802
## sexorientc_LGB 1.1614 0.8610 0.7846
## educb_highschool_grad 0.8732 1.1452 0.7148
## educc_college_bach 0.5464 1.8301 0.4089
## educd_college+ 0.3117 3.2085 0.2287
## unable_pay_utilitiesb_yes 1.3807 0.7243 1.2075
## insurance_statusW4b_yes_insurance 0.6533 1.5307 0.5780
## general_healthb_poor/bad 1.1627 0.8600 0.9612
## depressionW4b_yes 1.1911 0.8396 1.0141
## alcohol_day_permonthW4b_1or2days/week 0.9962 1.0038 0.8649
## alcohol_day_permonthW4c_3to5days/week 0.9945 1.0055 0.8262
## alcohol_day_permonthW4d_daily 1.2661 0.7898 0.9679
## leave_jobb_health 0.3074 3.2526 0.2005
## leave_jobc_other 0.3951 2.5309 0.3027
## leave_jobd_skip 0.4471 2.2368 0.3652
## job_satisfactionb_neutral 1.1182 0.8943 0.9555
## job_satisfactionc_dissatisfied 1.2206 0.8193 1.0142
## beaten_upW4b_yes 1.0195 0.9809 0.8387
## upper .95
## sexb_female 0.8054
## racethnicb-nhblack 1.5295
## racethnicc-hispanic 1.2637
## racethnicd-asian 1.0006
## racethnice-native_american 2.8481
## racethnicf-other 2.7408
## sexorientb_bisexual 1.9532
## sexorientc_LGB 1.7192
## educb_highschool_grad 1.0667
## educc_college_bach 0.7302
## educd_college+ 0.4248
## unable_pay_utilitiesb_yes 1.5787
## insurance_statusW4b_yes_insurance 0.7385
## general_healthb_poor/bad 1.4066
## depressionW4b_yes 1.3989
## alcohol_day_permonthW4b_1or2days/week 1.1474
## alcohol_day_permonthW4c_3to5days/week 1.1972
## alcohol_day_permonthW4d_daily 1.6563
## leave_jobb_health 0.4714
## leave_jobc_other 0.5158
## leave_jobd_skip 0.5473
## job_satisfactionb_neutral 1.3087
## job_satisfactionc_dissatisfied 1.4690
## beaten_upW4b_yes 1.2392
##
## Concordance= 0.671 (se = 0.008 )
## Likelihood ratio test= NA on 24 df, p=NA
## Wald test = 576 on 24 df, p=<2e-16
## Score (logrank) test = NA on 24 df, p=NAschoenfeld residuals
## Response sexb_female :
##
## Call:
## lm(formula = sexb_female ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.4797 -0.4563 -0.4427 0.5445 0.5634
##
## Coefficients:
## Estimate
## (Intercept) 0.0703658
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.0008703
## Std. Error
## (Intercept) 0.1467977
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0066452
## t value
## (Intercept) 0.479
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.131
## Pr(>|t|)
## (Intercept) 0.632
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.896
##
## Residual standard error: 0.4991 on 2107 degrees of freedom
## Multiple R-squared: 8.14e-06, Adjusted R-squared: -0.0004665
## F-statistic: 0.01715 on 1 and 2107 DF, p-value: 0.8958
##
##
## Response racethnicb-nhblack :
##
## Call:
## lm(formula = `racethnicb-nhblack` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2788 -0.2346 -0.2152 -0.1860 0.8233
##
## Coefficients:
## Estimate
## (Intercept) -0.172513
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.010885
## Std. Error
## (Intercept) 0.122445
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005543
## t value
## (Intercept) -1.409
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 1.964
## Pr(>|t|)
## (Intercept) 0.1590
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0497 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4163 on 2107 degrees of freedom
## Multiple R-squared: 0.001827, Adjusted R-squared: 0.001353
## F-statistic: 3.856 on 1 and 2107 DF, p-value: 0.04969
##
##
## Response racethnicc-hispanic :
##
## Call:
## lm(formula = `racethnicc-hispanic` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.07292 -0.06937 -0.06764 -0.06555 0.93857
##
## Coefficients:
## Estimate
## (Intercept) 0.041320
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.001475
## Std. Error
## (Intercept) 0.073982
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.003349
## t value
## (Intercept) 0.559
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.440
## Pr(>|t|)
## (Intercept) 0.577
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.660
##
## Residual standard error: 0.2515 on 2107 degrees of freedom
## Multiple R-squared: 9.201e-05, Adjusted R-squared: -0.0003826
## F-statistic: 0.1939 on 1 and 2107 DF, p-value: 0.6597
##
##
## Response racethnicd-asian :
##
## Call:
## lm(formula = `racethnicd-asian` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.05253 -0.04634 -0.04225 -0.03812 0.96746
##
## Coefficients:
## Estimate
## (Intercept) 0.081687
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.002760
## Std. Error
## (Intercept) 0.059440
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.002691
## t value
## (Intercept) 1.374
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.026
## Pr(>|t|)
## (Intercept) 0.169
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.305
##
## Residual standard error: 0.2021 on 2107 degrees of freedom
## Multiple R-squared: 0.0004991, Adjusted R-squared: 2.472e-05
## F-statistic: 1.052 on 1 and 2107 DF, p-value: 0.3051
##
##
## Response racethnice-native_american :
##
## Call:
## lm(formula = `racethnice-native_american` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.01323 -0.00879 -0.00754 -0.00610 0.99618
##
## Coefficients:
## Estimate
## (Intercept) -0.025103
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.001138
## Std. Error
## (Intercept) 0.025534
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.001156
## t value
## (Intercept) -0.983
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.985
## Pr(>|t|)
## (Intercept) 0.326
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.325
##
## Residual standard error: 0.0868 on 2107 degrees of freedom
## Multiple R-squared: 0.0004601, Adjusted R-squared: -1.426e-05
## F-statistic: 0.9699 on 1 and 2107 DF, p-value: 0.3248
##
##
## Response racethnicf-other :
##
## Call:
## lm(formula = `racethnicf-other` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.01087 -0.00923 -0.00837 -0.00787 0.99250
##
## Coefficients:
## Estimate
## (Intercept) -0.0062256
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0002495
## Std. Error
## (Intercept) 0.0270698
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0012254
## t value
## (Intercept) -0.230
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.204
## Pr(>|t|)
## (Intercept) 0.818
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.839
##
## Residual standard error: 0.09203 on 2107 degrees of freedom
## Multiple R-squared: 1.967e-05, Adjusted R-squared: -0.0004549
## F-statistic: 0.04144 on 1 and 2107 DF, p-value: 0.8387
##
##
## Response sexorientb_bisexual :
##
## Call:
## lm(formula = sexorientb_bisexual ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.04795 -0.02975 -0.02294 -0.01626 0.98966
##
## Coefficients:
## Estimate
## (Intercept) -0.131377
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005920
## Std. Error
## (Intercept) 0.044236
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.002002
## t value
## (Intercept) -2.970
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 2.956
## Pr(>|t|)
## (Intercept) 0.00301
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.00315
##
## (Intercept) **
## des2$variables$job.loss.age[des2$variables$job_transition == 1] **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1504 on 2107 degrees of freedom
## Multiple R-squared: 0.00413, Adjusted R-squared: 0.003658
## F-statistic: 8.739 on 1 and 2107 DF, p-value: 0.00315
##
##
## Response sexorientc_LGB :
##
## Call:
## lm(formula = sexorientc_LGB ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.02143 -0.01972 -0.01911 -0.01813 0.98432
##
## Coefficients:
## Estimate
## (Intercept) 0.012146
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.000571
## Std. Error
## (Intercept) 0.040148
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.001817
## t value
## (Intercept) 0.303
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.314
## Pr(>|t|)
## (Intercept) 0.762
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.753
##
## Residual standard error: 0.1365 on 2107 degrees of freedom
## Multiple R-squared: 4.685e-05, Adjusted R-squared: -0.0004277
## F-statistic: 0.09871 on 1 and 2107 DF, p-value: 0.7534
##
##
## Response educb_highschool_grad :
##
## Call:
## lm(formula = educb_highschool_grad ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.7332 -0.6565 0.3101 0.3323 0.3723
##
## Coefficients:
## Estimate
## (Intercept) -0.259099
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.011126
## Std. Error
## (Intercept) 0.137804
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.006238
## t value
## (Intercept) -1.880
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 1.784
## Pr(>|t|)
## (Intercept) 0.0602 .
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0746 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4685 on 2107 degrees of freedom
## Multiple R-squared: 0.001507, Adjusted R-squared: 0.001034
## F-statistic: 3.181 on 1 and 2107 DF, p-value: 0.07464
##
##
## Response educc_college_bach :
##
## Call:
## lm(formula = educc_college_bach ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2092 -0.1758 -0.1633 -0.1491 0.8709
##
## Coefficients:
## Estimate
## (Intercept) 0.157660
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.006446
## Std. Error
## (Intercept) 0.110409
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.004998
## t value
## (Intercept) 1.428
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.290
## Pr(>|t|)
## (Intercept) 0.153
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.197
##
## Residual standard error: 0.3753 on 2107 degrees of freedom
## Multiple R-squared: 0.0007889, Adjusted R-squared: 0.0003147
## F-statistic: 1.664 on 1 and 2107 DF, p-value: 0.1973
##
##
## Response educd_college+ :
##
## Call:
## lm(formula = `educd_college+` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.08448 -0.06766 -0.06264 -0.05482 0.96022
##
## Coefficients:
## Estimate
## (Intercept) 0.103726
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.004142
## Std. Error
## (Intercept) 0.071460
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.003235
## t value
## (Intercept) 1.452
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.280
## Pr(>|t|)
## (Intercept) 0.147
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.201
##
## Residual standard error: 0.2429 on 2107 degrees of freedom
## Multiple R-squared: 0.0007775, Adjusted R-squared: 0.0003033
## F-statistic: 1.64 on 1 and 2107 DF, p-value: 0.2005
##
##
## Response unable_pay_utilitiesb_yes :
##
## Call:
## lm(formula = unable_pay_utilitiesb_yes ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2395 -0.2209 -0.2062 -0.1819 0.8330
##
## Coefficients:
## Estimate
## (Intercept) -0.068553
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.002930
## Std. Error
## (Intercept) 0.120101
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005437
## t value
## (Intercept) -0.571
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.539
## Pr(>|t|)
## (Intercept) 0.568
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.590
##
## Residual standard error: 0.4083 on 2107 degrees of freedom
## Multiple R-squared: 0.0001378, Adjusted R-squared: -0.0003368
## F-statistic: 0.2904 on 1 and 2107 DF, p-value: 0.59
##
##
## Response insurance_statusW4b_yes_insurance :
##
## Call:
## lm(formula = insurance_statusW4b_yes_insurance ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.7311 -0.6576 0.2964 0.3254 0.3577
##
## Coefficients:
## Estimate
## (Intercept) -0.020331
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.001812
## Std. Error
## (Intercept) 0.136479
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.006178
## t value
## (Intercept) -0.149
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.293
## Pr(>|t|)
## (Intercept) 0.882
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.769
##
## Residual standard error: 0.464 on 2107 degrees of freedom
## Multiple R-squared: 4.081e-05, Adjusted R-squared: -0.0004338
## F-statistic: 0.086 on 1 and 2107 DF, p-value: 0.7694
##
##
## Response general_healthb_poor/bad :
##
## Call:
## lm(formula = `general_healthb_poor/bad` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.15369 -0.12821 -0.11494 -0.09932 0.91917
##
## Coefficients:
## Estimate
## (Intercept) 0.213729
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.009629
## Std. Error
## (Intercept) 0.094725
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.004288
## t value
## (Intercept) 2.256
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -2.246
## Pr(>|t|)
## (Intercept) 0.0242 *
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0248 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.322 on 2107 degrees of freedom
## Multiple R-squared: 0.002388, Adjusted R-squared: 0.001914
## F-statistic: 5.043 on 1 and 2107 DF, p-value: 0.02483
##
##
## Response depressionW4b_yes :
##
## Call:
## lm(formula = depressionW4b_yes ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2111 -0.1925 -0.1856 -0.1739 0.8414
##
## Coefficients:
## Estimate
## (Intercept) 0.128573
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.006133
## Std. Error
## (Intercept) 0.114995
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005206
## t value
## (Intercept) 1.118
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.178
## Pr(>|t|)
## (Intercept) 0.264
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.239
##
## Residual standard error: 0.3909 on 2107 degrees of freedom
## Multiple R-squared: 0.0006584, Adjusted R-squared: 0.0001841
## F-statistic: 1.388 on 1 and 2107 DF, p-value: 0.2389
##
##
## Response alcohol_day_permonthW4b_1or2days/week :
##
## Call:
## lm(formula = `alcohol_day_permonthW4b_1or2days/week` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2466 -0.2329 -0.2295 -0.2243 0.7790
##
## Coefficients:
## Estimate
## (Intercept) 0.031494
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.001655
## Std. Error
## (Intercept) 0.124150
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005620
## t value
## (Intercept) 0.254
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.294
## Pr(>|t|)
## (Intercept) 0.800
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.768
##
## Residual standard error: 0.4221 on 2107 degrees of freedom
## Multiple R-squared: 4.116e-05, Adjusted R-squared: -0.0004334
## F-statistic: 0.08672 on 1 and 2107 DF, p-value: 0.7684
##
##
## Response alcohol_day_permonthW4c_3to5days/week :
##
## Call:
## lm(formula = `alcohol_day_permonthW4c_3to5days/week` ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.1189 -0.1139 -0.1119 -0.1099 0.8936
##
## Coefficients:
## Estimate
## (Intercept) -0.0248842
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0009922
## Std. Error
## (Intercept) 0.0929568
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0042079
## t value
## (Intercept) -0.268
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.236
## Pr(>|t|)
## (Intercept) 0.789
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.814
##
## Residual standard error: 0.316 on 2107 degrees of freedom
## Multiple R-squared: 2.639e-05, Adjusted R-squared: -0.0004482
## F-statistic: 0.0556 on 1 and 2107 DF, p-value: 0.8136
##
##
## Response alcohol_day_permonthW4d_daily :
##
## Call:
## lm(formula = alcohol_day_permonthW4d_daily ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.04873 -0.04342 -0.04117 -0.03828 0.96741
##
## Coefficients:
## Estimate
## (Intercept) -0.028426
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.001054
## Std. Error
## (Intercept) 0.058230
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.002636
## t value
## (Intercept) -0.488
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.400
## Pr(>|t|)
## (Intercept) 0.625
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.689
##
## Residual standard error: 0.198 on 2107 degrees of freedom
## Multiple R-squared: 7.584e-05, Adjusted R-squared: -0.0003987
## F-statistic: 0.1598 on 1 and 2107 DF, p-value: 0.6894
##
##
## Response leave_jobb_health :
##
## Call:
## lm(formula = leave_jobb_health ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.04496 -0.03363 -0.02928 -0.02440 0.98007
##
## Coefficients:
## Estimate
## (Intercept) -0.072432
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.003427
## Std. Error
## (Intercept) 0.049684
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.002249
## t value
## (Intercept) -1.458
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 1.524
## Pr(>|t|)
## (Intercept) 0.145
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.128
##
## Residual standard error: 0.1689 on 2107 degrees of freedom
## Multiple R-squared: 0.001101, Adjusted R-squared: 0.0006266
## F-statistic: 2.322 on 1 and 2107 DF, p-value: 0.1277
##
##
## Response leave_jobc_other :
##
## Call:
## lm(formula = leave_jobc_other ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.06299 -0.05847 -0.05672 -0.05493 0.94709
##
## Coefficients:
## Estimate
## (Intercept) -0.0246842
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0009373
## Std. Error
## (Intercept) 0.0681578
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0030853
## t value
## (Intercept) -0.362
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.304
## Pr(>|t|)
## (Intercept) 0.717
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.761
##
## Residual standard error: 0.2317 on 2107 degrees of freedom
## Multiple R-squared: 4.38e-05, Adjusted R-squared: -0.0004308
## F-statistic: 0.09228 on 1 and 2107 DF, p-value: 0.7613
##
##
## Response leave_jobd_skip :
##
## Call:
## lm(formula = leave_jobd_skip ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.8888 0.1343 0.1567 0.1743 0.2172
##
## Coefficients:
## Estimate
## (Intercept) 0.182390
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.008485
## Std. Error
## (Intercept) 0.108554
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.004914
## t value
## (Intercept) 1.680
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.727
## Pr(>|t|)
## (Intercept) 0.0931 .
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.0844 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.369 on 2107 degrees of freedom
## Multiple R-squared: 0.001413, Adjusted R-squared: 0.0009392
## F-statistic: 2.982 on 1 and 2107 DF, p-value: 0.08436
##
##
## Response job_satisfactionb_neutral :
##
## Call:
## lm(formula = job_satisfactionb_neutral ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.2337 -0.2108 -0.2004 -0.1868 0.8213
##
## Coefficients:
## Estimate
## (Intercept) 0.164508
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.006806
## Std. Error
## (Intercept) 0.118590
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005368
## t value
## (Intercept) 1.387
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.268
## Pr(>|t|)
## (Intercept) 0.166
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.205
##
## Residual standard error: 0.4032 on 2107 degrees of freedom
## Multiple R-squared: 0.0007622, Adjusted R-squared: 0.0002879
## F-statistic: 1.607 on 1 and 2107 DF, p-value: 0.205
##
##
## Response job_satisfactionc_dissatisfied :
##
## Call:
## lm(formula = job_satisfactionc_dissatisfied ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.1502 -0.1290 -0.1217 -0.1129 0.8983
##
## Coefficients:
## Estimate
## (Intercept) -0.103593
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.005005
## Std. Error
## (Intercept) 0.096733
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.004379
## t value
## (Intercept) -1.071
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 1.143
## Pr(>|t|)
## (Intercept) 0.284
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.253
##
## Residual standard error: 0.3289 on 2107 degrees of freedom
## Multiple R-squared: 0.0006198, Adjusted R-squared: 0.0001455
## F-statistic: 1.307 on 1 and 2107 DF, p-value: 0.2531
##
##
## Response beaten_upW4b_yes :
##
## Call:
## lm(formula = beaten_upW4b_yes ~ des2$variables$job.loss.age[des2$variables$job_transition ==
## 1])
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.1378 -0.1242 -0.1138 -0.1071 0.9072
##
## Coefficients:
## Estimate
## (Intercept) 0.136825
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -0.005987
## Std. Error
## (Intercept) 0.094637
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.004284
## t value
## (Intercept) 1.446
## des2$variables$job.loss.age[des2$variables$job_transition == 1] -1.398
## Pr(>|t|)
## (Intercept) 0.148
## des2$variables$job.loss.age[des2$variables$job_transition == 1] 0.162
##
## Residual standard error: 0.3217 on 2107 degrees of freedom
## Multiple R-squared: 0.0009262, Adjusted R-squared: 0.000452
## F-statistic: 1.953 on 1 and 2107 DF, p-value: 0.1624Grambsch and Therneau test
## rho chisq p
## sexb_female -0.07999 2.97e+01 5.05e-08
## racethnicb-nhblack -0.01150 9.95e-01 3.18e-01
## racethnicc-hispanic 0.00549 1.48e-01 7.00e-01
## racethnicd-asian 0.05507 1.28e+01 3.42e-04
## racethnice-native_american 0.02766 4.38e+00 3.63e-02
## racethnicf-other -0.01340 9.56e-01 3.28e-01
## sexorientb_bisexual -0.01834 1.42e+00 2.34e-01
## sexorientc_LGB 0.01011 4.22e-01 5.16e-01
## educb_highschool_grad -0.06168 2.32e+01 1.45e-06
## educc_college_bach -0.07794 4.45e+01 2.50e-11
## educd_college+ -0.00112 6.66e-03 9.35e-01
## unable_pay_utilitiesb_yes 0.00919 3.75e-01 5.40e-01
## insurance_statusW4b_yes_insurance 0.03721 7.13e+00 7.57e-03
## general_healthb_poor/bad 0.01619 1.53e+00 2.17e-01
## depressionW4b_yes -0.00618 2.01e-01 6.54e-01
## alcohol_day_permonthW4b_1or2days/week -0.02137 2.15e+00 1.42e-01
## alcohol_day_permonthW4c_3to5days/week 0.02604 3.37e+00 6.64e-02
## alcohol_day_permonthW4d_daily -0.00814 3.21e-01 5.71e-01
## leave_jobb_health -0.04079 1.06e+01 1.15e-03
## leave_jobc_other -0.06359 1.65e+01 4.85e-05
## leave_jobd_skip -0.06195 1.77e+01 2.61e-05
## job_satisfactionb_neutral -0.07405 2.97e+01 5.03e-08
## job_satisfactionc_dissatisfied 0.03306 5.71e+00 1.68e-02
## beaten_upW4b_yes 0.03510 6.90e+00 8.63e-03
## GLOBAL NA 1.69e+02 9.33e-24Plot of residuals
###Martingale residuals
## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : pseudoinverse used at 0.99## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : neighborhood radius 2.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : reciprocal condition number 4.0323e-16## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : There are other near singularities as well. 4.0401## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : pseudoinverse used at 0.99## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : neighborhood radius 2.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : There are other near singularities as well. 4.0401## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : pseudoinverse used at 0.99## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : neighborhood radius 2.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : There are other near singularities as well. 4.0401## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : pseudoinverse used at 0.99## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : neighborhood radius 2.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : reciprocal condition number 0## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : There are other near singularities as well. 4.0401## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : pseudoinverse used at 0.99## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : neighborhood radius 2.01## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : reciprocal condition number 4.0323e-16## Warning in simpleLoess(y, x, w, span, degree = degree, parametric =
## FALSE, : There are other near singularities as well. 4.0401
###Stratification by sexual orientation
## Stratified 1 - level Cluster Sampling design (with replacement)
## With (130) clusters.
## svydesign(ids = ~psuscid, strata = ~region, weights = ~GSWGT4,
## data = addhealth, nest = T)## Call:
## svycoxph(formula = Surv(time = job.loss.age, event = job_transition) ~
## strata(sexorient) + sex + racethnic + educ + unable_pay_utilities +
## insurance_statusW4 + general_health + depressionW4 +
## alcohol_day_permonthW4 + leave_job + job_satisfaction +
## beaten_upW4, design = des2)
##
## n= 7101, number of events= 2109
##
## coef exp(coef) se(coef) z
## sexb_female -0.335703 0.714836 0.060583 -5.541
## racethnicb-nhblack 0.253031 1.287923 0.087930 2.878
## racethnicc-hispanic -0.001235 0.998766 0.120151 -0.010
## racethnicd-asian -0.279472 0.756183 0.143855 -1.943
## racethnice-native_american 0.392777 1.481088 0.332964 1.180
## racethnicf-other 0.356074 1.427714 0.332230 1.072
## educb_highschool_grad -0.135344 0.873416 0.101981 -1.327
## educc_college_bach -0.603440 0.546927 0.147813 -4.082
## educd_college+ -1.166681 0.311399 0.157663 -7.400
## unable_pay_utilitiesb_yes 0.321484 1.379173 0.068445 4.697
## insurance_statusW4b_yes_insurance -0.426206 0.652982 0.062538 -6.815
## general_healthb_poor/bad 0.149968 1.161797 0.097086 1.545
## depressionW4b_yes 0.175581 1.191939 0.082208 2.136
## alcohol_day_permonthW4b_1or2days/week -0.004832 0.995179 0.072246 -0.067
## alcohol_day_permonthW4c_3to5days/week -0.005481 0.994534 0.094793 -0.058
## alcohol_day_permonthW4d_daily 0.235276 1.265258 0.136874 1.719
## leave_jobb_health -1.179429 0.307454 0.218010 -5.410
## leave_jobc_other -0.929322 0.394821 0.136137 -6.826
## leave_jobd_skip -0.805227 0.446986 0.103213 -7.802
## job_satisfactionb_neutral 0.112813 1.119422 0.080290 1.405
## job_satisfactionc_dissatisfied 0.200945 1.222557 0.094657 2.123
## beaten_upW4b_yes 0.018515 1.018687 0.099776 0.186
## Pr(>|z|)
## sexb_female 3.00e-08 ***
## racethnicb-nhblack 0.00401 **
## racethnicc-hispanic 0.99180
## racethnicd-asian 0.05205 .
## racethnice-native_american 0.23814
## racethnicf-other 0.28382
## educb_highschool_grad 0.18446
## educc_college_bach 4.46e-05 ***
## educd_college+ 1.36e-13 ***
## unable_pay_utilitiesb_yes 2.64e-06 ***
## insurance_statusW4b_yes_insurance 9.41e-12 ***
## general_healthb_poor/bad 0.12242
## depressionW4b_yes 0.03269 *
## alcohol_day_permonthW4b_1or2days/week 0.94667
## alcohol_day_permonthW4c_3to5days/week 0.95389
## alcohol_day_permonthW4d_daily 0.08563 .
## leave_jobb_health 6.30e-08 ***
## leave_jobc_other 8.71e-12 ***
## leave_jobd_skip 6.11e-15 ***
## job_satisfactionb_neutral 0.16000
## job_satisfactionc_dissatisfied 0.03376 *
## beaten_upW4b_yes 0.85279
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## exp(coef) exp(-coef) lower .95
## sexb_female 0.7148 1.3989 0.6348
## racethnicb-nhblack 1.2879 0.7764 1.0840
## racethnicc-hispanic 0.9988 1.0012 0.7892
## racethnicd-asian 0.7562 1.3224 0.5704
## racethnice-native_american 1.4811 0.6752 0.7712
## racethnicf-other 1.4277 0.7004 0.7445
## educb_highschool_grad 0.8734 1.1449 0.7152
## educc_college_bach 0.5469 1.8284 0.4094
## educd_college+ 0.3114 3.2113 0.2286
## unable_pay_utilitiesb_yes 1.3792 0.7251 1.2060
## insurance_statusW4b_yes_insurance 0.6530 1.5314 0.5777
## general_healthb_poor/bad 1.1618 0.8607 0.9605
## depressionW4b_yes 1.1919 0.8390 1.0146
## alcohol_day_permonthW4b_1or2days/week 0.9952 1.0048 0.8638
## alcohol_day_permonthW4c_3to5days/week 0.9945 1.0055 0.8259
## alcohol_day_permonthW4d_daily 1.2653 0.7904 0.9675
## leave_jobb_health 0.3075 3.2525 0.2005
## leave_jobc_other 0.3948 2.5328 0.3024
## leave_jobd_skip 0.4470 2.2372 0.3651
## job_satisfactionb_neutral 1.1194 0.8933 0.9564
## job_satisfactionc_dissatisfied 1.2226 0.8180 1.0155
## beaten_upW4b_yes 1.0187 0.9817 0.8377
## upper .95
## sexb_female 0.8050
## racethnicb-nhblack 1.5302
## racethnicc-hispanic 1.2640
## racethnicd-asian 1.0025
## racethnice-native_american 2.8445
## racethnicf-other 2.7380
## educb_highschool_grad 1.0667
## educc_college_bach 0.7307
## educd_college+ 0.4242
## unable_pay_utilitiesb_yes 1.5772
## insurance_statusW4b_yes_insurance 0.7381
## general_healthb_poor/bad 1.4053
## depressionW4b_yes 1.4003
## alcohol_day_permonthW4b_1or2days/week 1.1466
## alcohol_day_permonthW4c_3to5days/week 1.1976
## alcohol_day_permonthW4d_daily 1.6546
## leave_jobb_health 0.4714
## leave_jobc_other 0.5156
## leave_jobd_skip 0.5472
## job_satisfactionb_neutral 1.3102
## job_satisfactionc_dissatisfied 1.4718
## beaten_upW4b_yes 1.2387
##
## Concordance= 0.672 (se = 0.008 )
## Likelihood ratio test= NA on 22 df, p=NA
## Wald test = 550.3 on 22 df, p=<2e-16
## Score (logrank) test = NA on 22 df, p=NA