D18=read.csv("output18_update.csv") D12=read.csv("output12_update.csv") D04=read.csv("output04_update.csv") D95=read.csv("output95_update.csv") d18a=lm(LNWAGE~FEMALE+NONWHITE+UNION+poly(EDUC,2)+poly(AGE,2),data=D18) summary(d18a) Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 3.130487 0.006665 469.717 < 2e-16 *** FEMALETRUE -0.191345 0.008804 -21.733 < 2e-16 *** NONWHITETRUE -0.074081 0.010993 -6.739 1.67e-11 *** UNIONTRUE 0.117967 0.014530 8.119 5.17e-16 *** poly(EDUC, 2)1 26.730573 0.486688 54.923 < 2e-16 *** poly(EDUC, 2)2 5.630225 0.479598 11.739 < 2e-16 *** poly(AGE, 2)1 11.740709 0.483522 24.282 < 2e-16 *** poly(AGE, 2)2 -12.259723 0.484537 -25.302 < 2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4781 on 11858 degrees of freedom Multiple R-squared: 0.3303, Adjusted R-squared: 0.3299 F-statistic: 835.7 on 7 and 11858 DF, p-value: < 2.2e-16 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 3.130487 0.006665 469.717 < 2e-16 *** FEMALETRUE -0.191345 0.008804 -21.733 < 2e-16 *** NONWHITETRUE -0.074081 0.010993 -6.739 1.67e-11 *** UNIONTRUE 0.117967 0.014530 8.119 5.17e-16 *** poly(EDUC, 2)1 26.730573 0.486688 54.923 < 2e-16 *** poly(EDUC, 2)2 5.630225 0.479598 11.739 < 2e-16 *** poly(AGE, 2)1 11.740709 0.483522 24.282 < 2e-16 *** poly(AGE, 2)2 -12.259723 0.484537 -25.302 < 2e-16 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4781 on 11858 degrees of freedom Multiple R-squared: 0.3303, Adjusted R-squared: 0.3299 F-statistic: 835.7 on 7 and 11858 DF, p-value: < 2.2e-16 d12a=lm(LNWAGE~FEMALE+NONWHITE+UNION+poly(EDUC,2)+poly(AGE,2),data=D12) summary(d12a) Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 2.92094 0.02204 132.545 < 2e-16 *** FEMALE -0.17192 0.02952 -5.824 7.33e-09 *** NONWHITE -0.04127 0.03914 -1.054 0.291887 UNION 0.15802 0.04509 3.504 0.000474 *** poly(EDUC, 2)1 8.66250 0.51761 16.736 < 2e-16 *** poly(EDUC, 2)2 1.97589 0.51723 3.820 0.000140 *** poly(AGE, 2)1 4.70849 0.51753 9.098 < 2e-16 *** poly(AGE, 2)2 -3.03759 0.51761 -5.868 5.66e-09 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.5142 on 1221 degrees of freedom Multiple R-squared: 0.297, Adjusted R-squared: 0.293 F-statistic: 73.71 on 7 and 1221 DF, p-value: < 2.2e-16 d04a=lm(LNWAGE~FEMALE+NONWHITE+UNION+poly(EDUC,2)+poly(AGE,2),data=D04) summary(d04a) Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 2.79702 0.01937 144.378 < 2e-16 *** FEMALE -0.24930 0.02507 -9.944 < 2e-16 *** NONWHITE -0.10449 0.03097 -3.374 0.000755 *** UNION 0.08214 0.03797 2.163 0.030658 * poly(EDUC, 2)1 10.32960 0.55800 18.512 < 2e-16 *** poly(EDUC, 2)2 2.33385 0.55182 4.229 2.45e-05 *** poly(AGE, 2)1 6.69344 0.55670 12.024 < 2e-16 *** poly(AGE, 2)2 -3.76886 0.55360 -6.808 1.32e-11 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.5477 on 1933 degrees of freedom Multiple R-squared: 0.2936, Adjusted R-squared: 0.2911 F-statistic: 114.8 on 7 and 1933 DF, p-value: < 2.2e-16 d95a=lm(LNWAGE~FEMALE+NONWHITE+UNION+poly(EDUC,2)+poly(AGE,2),data=D95) summary(d95a) Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 2.43980 0.01943 125.578 < 2e-16 *** FEMALE -0.23580 0.02543 -9.274 < 2e-16 *** NONWHITE -0.09555 0.03492 -2.736 0.0063 ** UNION 0.20230 0.03456 5.854 6.07e-09 *** poly(EDUC, 2)1 7.23428 0.46267 15.636 < 2e-16 *** poly(EDUC, 2)2 0.95478 0.46062 2.073 0.0384 * poly(AGE, 2)1 5.32060 0.46458 11.453 < 2e-16 *** poly(AGE, 2)2 -3.72663 0.46371 -8.037 2.04e-15 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4564 on 1315 degrees of freedom Multiple R-squared: 0.3426, Adjusted R-squared: 0.3391 F-statistic: 97.91 on 7 and 1315 DF, p-value: < 2.2e-16 coefs FEMALE NONWHITE UNION R^2 95 -0.236 -0.096 0.202 0.3391 4 -0.249 -0.104 0.082 0.2911 12 -0.172 -0.041 0.158 0.293 18 -0.191 -0.074 0.118 0.3299 1 exception means FEMALE NONWHITE UNION EDUC AGE WAGE 95 0.492 0.155 0.163 13.1 38.3 12.2 4 0.514 0.205 0.125 13.3 39.6 17.1 12 0.471 0.171 0.122 13.7 41.6 20.7 18 0.489 0.201 0.103 13.9 42.2 24.9 declines increases increases increases > out18b=lm(LNWAGE~.-MJIND-MJOCC-GEDIV-PTOT-WAGE,data=D18) > summary(out18b) Call: lm(formula = LNWAGE ~ . - MJIND - MJOCC - GEDIV - PTOT - WAGE, data = D18) Residuals: Min 1Q Median 3Q Max -2.65937 -0.32376 -0.02272 0.31439 3.15747 Coefficients: (1 not defined because of singularities) Estimate Std. Error t value Pr(>|t|) (Intercept) 1.3451740 0.0348057 38.648 < 2e-16 *** AGE 0.0078217 0.0003277 23.870 < 2e-16 *** FEMALETRUE -0.1916391 0.0091993 -20.832 < 2e-16 *** EDUC 0.0855971 0.0016607 51.544 < 2e-16 *** VETTRUE -0.0318109 0.0199775 -1.592 0.1113 PRCITSHPTRUE 0.0417761 0.0172440 2.423 0.0154 * UNIONTRUE 0.1189357 0.0149699 7.945 2.12e-15 *** NOEMP 0.0173122 0.0023442 7.385 1.63e-13 *** PERLIS 0.0744769 0.0066634 11.177 < 2e-16 *** HEA -0.0430301 0.0051206 -8.403 < 2e-16 *** EXPER NA NA NA NA NONWHITETRUE -0.0614739 0.0113818 -5.401 6.75e-08 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4882 on 11855 degrees of freedom Multiple R-squared: 0.3019, Adjusted R-squared: 0.3013 F-statistic: 512.6 on 10 and 11855 DF, p-value: < 2.2e-16 (A) You are a vet, decrease, not sig (B) You are not a U.S. citizen, increase, sig (C) You work for a large company, increase, sig (D) You live well above the poverty line, increase, sig (E) Your health is poor. decrease, sig > goI=function(i){return(t.test(WAGE~FEMALE,data=D18,subset=(MJIND==i))$p.value)} > sapply(1:13,goI) [1] 5.616824e-01 8.461865e-01 9.843946e-01 8.166810e-09 1.090548e-12 [6] 4.328979e-01 8.869863e-03 1.444133e-04 9.435816e-10 1.135944e-10 [11] 7.171616e-04 1.947271e-03 1.360336e-02 V 4 .Manufacturing V 5 .Wholesale and retail trade V 7 .Information V 8 .Financial activities V 9 .Professional and business V .services V 10 .Educational and health services V 11 .Leisure and hospitality V 12 .Other services V 13 .Public administration > goO=function(i){return(t.test(WAGE~FEMALE,data=D18,subset=(MJOCC==i))$p.value)} > sapply(1:10,goO) [1] 1.444260e-16 8.496802e-18 4.694570e-16 1.490641e-19 1.330472e-01 [6] 6.770066e-01 5.462173e-01 1.729074e-03 1.020917e-05 7.896522e-01 V 1 .Management, business, and V .financial occupations V 2 .Professional and related V .occupations V 3 .Service occupations V 4 .Sales and related occupations V 8 .Installation, maintenance, V .and repair occupations V 9 .Production occupations > goG=function(i){return(t.test(WAGE~FEMALE,data=D18,subset=(GEDIV==i))$p.value)} > sapply(1:9,goG) [1] 2.043233e-07 6.414602e-07 8.006170e-05 1.016574e-08 5.674322e-08 [6] 8.971168e-03 1.691799e-13 2.553467e-10 6.128467e-02 not Pacific --- > goE=function(i){return(t.test(EDUC~FEMALE,data=D18,subset=(MJIND==i))$p.value)} > sapply(1:13,goE) [1] 3.305182e-01 8.187015e-01 4.077976e-09 3.799356e-01 3.467558e-01 [6] 1.289497e-01 4.977004e-01 1.171089e-04 8.987859e-03 1.135469e-13 [11] 5.999477e-01 1.357337e-01 2.558012e-02 V 3 .Construction V 8 .Financial activities V 9 .Professional and business V .services V 10 .Educational and health services V 13 .Public administration > goA=function(i){return(t.test(AGE~FEMALE,data=D18,subset=(MJOCC==i))$p.value)} > sapply(1:10,goA) [1] 3.004710e-01 1.337952e-01 5.459710e-02 1.470982e-03 2.333868e-11 [6] 1.982111e-01 5.143337e-01 4.720968e-01 2.229149e-01 3.153447e-01 D A_MJOCC 2 211 (00:11) Major occupation recode U A_CLSWKR = 1-7 V 3 .Service occupations almost V 4 .Sales and related occupations V 5 .Office and administrative V .support occupations > D18a=D18 > D18a$MJIND=as.factor(D18a$MJIND) > D18a$MJOCC=as.factor(D18a$MJOCC) > D18a$GEDIV=as.factor(D18a$GEDIV) > > a18=lm(LNWAGE~FEMALE*MJOCC+FEMALE*MJIND+FEMALE*GEDIV+NONWHITE+UNION+poly(AGE,2)+ + poly(EDUC,2),data=D18a) > summary(a18) Call: lm(formula = LNWAGE ~ FEMALE * MJOCC + FEMALE * MJIND + FEMALE * GEDIV + NONWHITE + UNION + poly(AGE, 2) + poly(EDUC, 2), data = D18a) Residuals: Min 1Q Median 3Q Max -2.43824 -0.28667 -0.01159 0.27446 2.98872 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 3.352209 0.071531 46.864 < 2e-16 *** FEMALETRUE -0.273565 0.129316 -2.115 0.03441 * MJOCC2 -0.082982 0.020733 -4.002 6.31e-05 *** MJOCC3 -0.407310 0.023935 -17.017 < 2e-16 *** MJOCC4 -0.195264 0.027357 -7.138 1.01e-12 *** MJOCC5 -0.407161 0.027431 -14.843 < 2e-16 *** MJOCC6 -0.450386 0.082586 -5.454 5.04e-08 *** MJOCC7 -0.278742 0.032170 -8.665 < 2e-16 *** MJOCC8 -0.173865 0.029503 -5.893 3.89e-09 *** MJOCC9 -0.333683 0.028556 -11.685 < 2e-16 *** MJOCC10 -0.428425 0.025689 -16.678 < 2e-16 *** MJIND2 0.227257 0.085938 2.644 0.00819 ** MJIND3 0.111077 0.072973 1.522 0.12799 MJIND4 0.124484 0.070067 1.777 0.07565 . MJIND5 -0.015921 0.070077 -0.227 0.82027 MJIND6 0.169936 0.071250 2.385 0.01709 * MJIND7 0.071310 0.077788 0.917 0.35931 MJIND8 0.150055 0.071973 2.085 0.03710 * MJIND9 0.099925 0.070059 1.426 0.15381 MJIND10 -0.091875 0.070293 -1.307 0.19123 MJIND11 -0.094346 0.071104 -1.327 0.18457 MJIND12 -0.034795 0.074210 -0.469 0.63916 MJIND13 0.126543 0.071380 1.773 0.07628 . GEDIV2 -0.030706 0.028341 -1.083 0.27863 GEDIV3 -0.085056 0.026869 -3.166 0.00155 ** GEDIV4 -0.075230 0.027920 -2.694 0.00706 ** GEDIV5 -0.060850 0.024447 -2.489 0.01282 * GEDIV6 -0.145736 0.030234 -4.820 1.45e-06 *** GEDIV7 -0.066663 0.026655 -2.501 0.01240 * GEDIV8 -0.055245 0.026380 -2.094 0.03627 * GEDIV9 0.033885 0.025148 1.347 0.17787 NONWHITETRUE -0.065098 0.010592 -6.146 8.20e-10 *** UNIONTRUE 0.117455 0.014362 8.178 3.18e-16 *** poly(AGE, 2)1 9.835976 0.467566 21.037 < 2e-16 *** poly(AGE, 2)2 -9.663789 0.465747 -20.749 < 2e-16 *** poly(EDUC, 2)1 19.178458 0.556179 34.483 < 2e-16 *** poly(EDUC, 2)2 4.180881 0.464768 8.996 < 2e-16 *** FEMALETRUE:MJOCC2 -0.028041 0.028835 -0.972 0.33085 FEMALETRUE:MJOCC3 0.005440 0.031931 0.170 0.86472 FEMALETRUE:MJOCC4 -0.143633 0.039093 -3.674 0.00024 *** FEMALETRUE:MJOCC5 0.141569 0.034088 4.153 3.30e-05 *** FEMALETRUE:MJOCC6 0.053262 0.148654 0.358 0.72013 FEMALETRUE:MJOCC7 0.180311 0.116735 1.545 0.12247 FEMALETRUE:MJOCC8 -0.193020 0.111306 -1.734 0.08292 . FEMALETRUE:MJOCC9 0.015346 0.048968 0.313 0.75399 FEMALETRUE:MJOCC10 0.092039 0.052545 1.752 0.07986 . FEMALETRUE:MJIND2 0.144261 0.197054 0.732 0.46413 FEMALETRUE:MJIND3 0.092186 0.139743 0.660 0.50947 FEMALETRUE:MJIND4 0.038798 0.128209 0.303 0.76219 FEMALETRUE:MJIND5 0.068680 0.127903 0.537 0.59130 FEMALETRUE:MJIND6 0.080478 0.132559 0.607 0.54379 FEMALETRUE:MJIND7 0.073621 0.138057 0.533 0.59386 FEMALETRUE:MJIND8 0.040430 0.129170 0.313 0.75429 FEMALETRUE:MJIND9 0.049028 0.127784 0.384 0.70123 FEMALETRUE:MJIND10 0.161022 0.126984 1.268 0.20480 FEMALETRUE:MJIND11 0.145667 0.128390 1.135 0.25658 FEMALETRUE:MJIND12 0.087610 0.131529 0.666 0.50537 FEMALETRUE:MJIND13 0.063283 0.129462 0.489 0.62498 FEMALETRUE:GEDIV2 0.037494 0.040007 0.937 0.34869 FEMALETRUE:GEDIV3 0.059636 0.037954 1.571 0.11615 FEMALETRUE:GEDIV4 -0.005997 0.039315 -0.153 0.87877 FEMALETRUE:GEDIV5 0.056156 0.034196 1.642 0.10058 FEMALETRUE:GEDIV6 0.018752 0.042966 0.436 0.66252 FEMALETRUE:GEDIV7 -0.037916 0.037758 -1.004 0.31531 FEMALETRUE:GEDIV8 0.009748 0.037165 0.262 0.79310 FEMALETRUE:GEDIV9 0.059095 0.035641 1.658 0.09733 . --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4495 on 11800 degrees of freedom Multiple R-squared: 0.4109, Adjusted R-squared: 0.4077 F-statistic: 126.6 on 65 and 11800 DF, p-value: < 2.2e-16 (A) nonwhite, female union member and {\tt MJIND11, MJOCC4, GEDIV9} 3.352209-0.273565-0.065098+0.117455-0.094346-0.19526+0.033885=2.87528 interactions: 0.145667-0.143633+0.059095=.061129 net: 2.936409 (B) white, male, non-union member, and {\tt MJIND1, MJOCC1, GEDIV1} 3.352209 (C) white, female non-union member and {\tt MJIND1, MJOCC1, GEDIV1} 3.352209-0.273565=3.078644 MJOCC? compared to MJOCC1=1 .Management, business, and .financial occupations max MJIND: 2 .Mining max female MJOCC: 7 .Construction and extraction .occupations max wage female: IND2 for females: 0.371518 2 .Mining (note very small female coverage) no positive OCC (use 1) 1 .Management, business, and .financial occupations GEDIV9: 0.092980 Pacific 3.352209-0.273565+0.371518+0.092980=3.543142 max wage male: management, mining, pacific 3.352209+0.227257+0.033885=3.613351 #5 Etest showed min at j=6 but result muted and variable delta2(i,j) variable: max in recent lowest wages in cubic fit: EDUC=5 or so my delta2(6,6) 1.0137722 1.0342746 > d95a=lm(LNWAGE~FEMALE+NONWHITE+UNION+poly(EDUC,2)+poly(AGE,2),data=D95) > summary(d95a) Call: lm(formula = LNWAGE ~ FEMALE + NONWHITE + UNION + poly(EDUC, 2) + poly(AGE, 2), data = D95) Residuals: Min 1Q Median 3Q Max -1.86801 -0.29041 0.00051 0.28928 1.77893 Coefficients: Estimate Std. Error t value Pr(>|t|) (Intercept) 2.43980 0.01943 125.578 < 2e-16 *** FEMALE -0.23580 0.02543 -9.274 < 2e-16 *** NONWHITE -0.09555 0.03492 -2.736 0.0063 ** UNION 0.20230 0.03456 5.854 6.07e-09 *** poly(EDUC, 2)1 7.23428 0.46267 15.636 < 2e-16 *** poly(EDUC, 2)2 0.95478 0.46062 2.073 0.0384 * poly(AGE, 2)1 5.32060 0.46458 11.453 < 2e-16 *** poly(AGE, 2)2 -3.72663 0.46371 -8.037 2.04e-15 *** --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 Residual standard error: 0.4564 on 1315 degrees of freedom Multiple R-squared: 0.3426, Adjusted R-squared: 0.3391 F-statistic: 97.91 on 7 and 1315 DF, p-value: < 2.2e-16 > boot(D95,bootit,1000) ORDINARY NONPARAMETRIC BOOTSTRAP Call: boot(data = D95, statistic = bootit, R = 1000) Bootstrap Statistics : original bias std. error from LM t1* 2.43980006 -1.226523e-03 0.01942354 0.01943 t2* -0.23579901 6.097309e-05 0.02452864 0.02543 t3* -0.09555181 2.310239e-03 0.03527288 0.03492 t4* 0.20230413 1.427704e-03 0.03056979 0.03456 t5* 7.23428381 3.005358e-02 0.51876494 0.46267 biggest diff t6* 0.95477926 -1.579043e-02 0.52381631 0.46062 t7* 5.32059684 1.128427e-03 0.45768966 0.46458 t8* -3.72662984 1.772920e-02 0.45187706 0.46371