OPTIONS NOCENTER formchar = "|----|+|---+="; FILENAME IN1 'C:\DATA\ROBIN2.dat'; DATA ONE; infile in1; INPUT id quit int time helmert1 helmert2 helmert3 h1time h2time h3time racew tv manual; /* categorical version of time */ TIMEC = TIME+1; /* have time so that post-int = 0, 6 mo = 1, 12 mo = 2, 24 mo = 4 */ if time=3 then time=4; PROC GENMOD DESCENDING; CLASS ID TIMEC; MODEL QUIT = TIME TIME*TIME HELMERT1 HELMERT2 HELMERT3 RACEW / LINK=LOGIT DIST=BIN; REPEATED SUBJECT=ID / WITHIN=TIMEC CORRW TYPE=UN; TITLE 'Unspecified Working Correlation'; PROC GENMOD DESCENDING; CLASS ID TIMEC; MAKE 'GEEEMPPEST' OUT = EXEST1; MAKE 'GEERCOV' OUT = EXCOV1; MODEL QUIT = TIME TIME*TIME HELMERT1 HELMERT2 HELMERT3 RACEW H1TIME H2TIME H3TIME / LINK=LOGIT DIST=BIN; REPEATED SUBJECT=ID / WITHIN=TIMEC CORRW TYPE=UN COVB; TITLE 'Unspecified Working Correlation'; PROC GENMOD DESCENDING data=one; CLASS ID TIMEC; MAKE 'GEEEMPPEST' OUT = EXEST; MAKE 'GEERCOV' OUT = EXCOV; MODEL QUIT = TIME TIME*TIME HELMERT1 HELMERT2 HELMERT3 RACEW H1TIME H2TIME H3TIME H1TIME*TIME H2TIME*TIME H3TIME*TIME / LINK=LOGIT DIST=BIN; REPEATED SUBJECT=ID / WITHIN=TIMEC CORRW TYPE=UN COVB; TITLE 'Unspecified Working Correlation'; RUN; DATA EXE; SET EXEST; PROC PRINT; TITLE 'UN Model with interactions - Parameter estimates'; RUN; DATA EXC; SET EXCOV; PROC PRINT; TITLE 'UN Model with interactions - Variance-covariances of estimates'; RUN; PROC IML; TITLE 'Using IML for multi-parameter Wald tests'; print 'Unspecified GEE model with LIN and QUAD group interactions'; USE EXEST; READ ALL INTO EXESTM; USE EXCOV; READ ALL INTO EXCOVM; PRINT EXESTM [FORMAT=8.4]; PRINT EXCOVM [FORMAT=8.4]; EXESTB = EXESTM[1:13,1]; /* select only the regression estimates */ PRINT EXESTB [FORMAT=8.4]; /* calculate the individual Wald statistics to double-check things */ wald = exestb / sqrt(vecdiag(excovm)); print wald [format=8.4]; /* group by time (quad) interaction */ w1 = {0 0 0 0 0 0 0 0 0 0 1 0 0, /* h1 by time2*/ 0 0 0 0 0 0 0 0 0 0 0 1 0, /* h2 by time2*/ 0 0 0 0 0 0 0 0 0 0 0 0 1}; /* h3 by time2*/ wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,3); print 'Wald statistic - group by QUAD time - 3 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; /* group by time (linear & quad) interaction */ w1 = {0 0 0 0 0 0 0 1 0 0 0 0 0, /* h1 by time */ 0 0 0 0 0 0 0 0 1 0 0 0 0, /* h2 by time */ 0 0 0 0 0 0 0 0 0 1 0 0 0, /* h3 by time */ 0 0 0 0 0 0 0 0 0 0 1 0 0, /* h1 by time2*/ 0 0 0 0 0 0 0 0 0 0 0 1 0, /* h2 by time2*/ 0 0 0 0 0 0 0 0 0 0 0 0 1}; /* h3 by time2*/ wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,6); print 'Wald statistic - group by LIN & QUAD time - 6 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; print 'Unspecified GEE model with LIN group interactions'; USE EXEST1; READ ALL INTO EXESTM; USE EXCOV1; READ ALL INTO EXCOVM; PRINT EXESTM [FORMAT=8.4]; PRINT EXCOVM [FORMAT=8.4]; EXESTB = EXESTM[1:10,1]; /* select only the regression estimates */ PRINT EXESTB [FORMAT=8.4]; /* calculate the individual Wald statistics to double-check things */ wald = exestb / sqrt(vecdiag(excovm)); print wald [format=8.4]; /* group by LIN time interaction */ w1 = {0 0 0 0 0 0 0 1 0 0, /* h1 by time */ 0 0 0 0 0 0 0 0 1 0, /* h2 by time */ 0 0 0 0 0 0 0 0 0 1}; /* h3 by time */ wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,3); print 'Wald statistic - group by LIN time - 3 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; /* H1 contrast at the final timepoint */ w1 = {0 0 0 1 0 0 0 4 0 0}; /* h1 at time=4 */ w1e = w1*EXESTB; wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,1); print 'H1 estimate at time=4' , w1e [FORMAT=8.3]; print 'Wald statistic - H1 at time=4 - 1 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; /* H2 contrast at the final timepoint */ w1 = {0 0 0 0 1 0 0 0 4 0}; /* h2 at time=4 */ w1e = w1*EXESTB; wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,1); print 'H2 estimate at time=4' , w1e [FORMAT=8.3]; print 'Wald statistic - H2 at time=4 - 1 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; /* H3 contrast at the final timepoint */ w1 = {0 0 0 0 0 1 0 0 0 4}; /* h3 at time=4 */ w1e = w1*EXESTB; wald1 = (T(EXESTB)*T(w1))*INV(w1*EXCOVM*(T(w1)))*(w1*EXESTB); wald1sig = 1 - probchi(wald1,1); print 'H3 estimate at time=4' , w1e [FORMAT=8.3]; print 'Wald statistic - H3 at time=4 - 1 df', wald1 [FORMAT=8.3]; print 'Wald statistic p-value ', wald1sig [FORMAT=8.3]; Unspecified Working Correlation The GENMOD Procedure Model Information Data Set WORK.ONE Distribution Binomial Link Function Logit Dependent Variable quit Observations Used 1744 Probability Modeled Pr( quit = 1 ) Class Level Information Class Levels Values id 489 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 ... TIMEC 4 1 2 3 4 Response Profile Ordered Ordered Level Value Count 1 1 383 2 0 1361 Parameter Information Parameter Effect Prm1 Intercept Prm2 time Prm3 time*time Prm4 helmert1 Prm5 helmert2 Prm6 helmert3 Prm7 racew Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 1737 1776.6303 1.0228 Scaled Deviance 1737 1776.6303 1.0228 Pearson Chi-Square 1737 1754.0313 1.0098 Scaled Pearson X2 1737 1754.0313 1.0098 Log Likelihood -888.3151 Algorithm converged. Analysis Of Initial Parameter Estimates Standard Wald 95% Confidence Chi- Parameter DF Estimate Error Limits Square Pr > ChiSq Intercept 1 -0.9832 0.1053 -1.1895 -0.7769 87.26 <.0001 time 1 -0.6746 0.1403 -0.9496 -0.3997 23.12 <.0001 time*time 1 0.1419 0.0338 0.0757 0.2080 17.65 <.0001 helmert1 1 0.4893 0.1214 0.2514 0.7272 16.25 <.0001 helmert2 1 0.2577 0.0906 0.0801 0.4353 8.08 0.0045 helmert3 1 0.1950 0.0886 0.0213 0.3687 4.84 0.0278 racew 1 0.4275 0.1404 0.1523 0.7026 9.27 0.0023 Scale 0 1.0000 0.0000 1.0000 1.0000 NOTE: The scale parameter was held fixed. GEE Model Information Correlation Structure Unstructured Within-Subject Effect TIMEC (4 levels) Subject Effect id (489 levels) Number of Clusters 489 Correlation Matrix Dimension 4 Maximum Cluster Size 4 Minimum Cluster Size 1 Algorithm converged. Working Correlation Matrix Col1 Col2 Col3 Col4 Row1 1.0000 0.2834 0.3173 0.2474 Row2 0.2834 1.0000 0.4410 0.2770 Row3 0.3173 0.4410 1.0000 0.5365 Row4 0.2474 0.2770 0.5365 1.0000 Analysis Of GEE Parameter Estimates Empirical Standard Error Estimates Standard 95% Confidence Parameter Estimate Error Limits Z Pr > |Z| Intercept -0.9990 0.1119 -1.2182 -0.7798 -8.93 <.0001 time -0.6330 0.1262 -0.8804 -0.3857 -5.02 <.0001 time*time 0.1317 0.0288 0.0752 0.1881 4.57 <.0001 helmert1 0.5826 0.1697 0.2500 0.9152 3.43 0.0006 helmert2 0.2881 0.1213 0.0503 0.5259 2.37 0.0176 helmert3 0.2019 0.1194 -0.0321 0.4359 1.69 0.0909 racew 0.3583 0.2003 -0.0343 0.7509 1.79 0.0736 Unspecified Working Correlation The GENMOD Procedure .... GEE Model Information Correlation Structure Unstructured Within-Subject Effect TIMEC (4 levels) Subject Effect id (489 levels) Number of Clusters 489 Correlation Matrix Dimension 4 Maximum Cluster Size 4 Minimum Cluster Size 1 Covariance Matrix (Model-Based) Prm1 Prm2 Prm3 Prm4 Prm5 Prm1 0.01315 -0.005725 0.0009194 -0.006005 0.001306 Prm2 -0.005725 0.01402 -0.003036 0.001753 -0.001075 Prm3 0.0009194 -0.003036 0.0007265 -0.000173 0.0001320 Prm4 -0.006005 0.001753 -0.000173 0.04205 0.003554 Prm5 0.001306 -0.001075 0.0001320 0.003554 0.02062 Prm6 -0.001352 0.0004021 -0.000017 -0.001965 -0.002677 Prm7 -0.008784 -0.000173 0.0000437 0.003896 0.006729 Prm8 0.001471 -0.001198 0.0001625 -0.01112 -0.000699 Prm9 -0.000712 0.0007949 -0.000074 -0.000743 -0.005073 Prm10 0.0004558 -0.000511 0.0000545 0.0004457 0.0005925 Covariance Matrix (Model-Based) Prm6 Prm7 Prm8 Prm9 Prm10 Prm1 -0.001352 -0.008784 0.001471 -0.000712 0.0004558 Prm2 0.0004021 -0.000173 -0.001198 0.0007949 -0.000511 Prm3 -0.000017 0.0000437 0.0001625 -0.000074 0.0000545 Prm4 -0.001965 0.003896 -0.01112 -0.000743 0.0004457 Prm5 -0.002677 0.006729 -0.000699 -0.005073 0.0005925 Prm6 0.02030 -0.001709 0.0004415 0.0005939 -0.005235 Prm7 -0.001709 0.03693 0.0001259 -0.000142 0.0000202 Prm8 0.0004415 0.0001259 0.008633 0.0006586 -0.000389 Prm9 0.0005939 -0.000142 0.0006586 0.004734 -0.000519 Prm10 -0.005235 0.0000202 -0.000389 -0.000519 0.004864 Covariance Matrix (Empirical) Prm1 Prm2 Prm3 Prm4 Prm5 Prm1 0.01357 -0.007012 0.001183 -0.007330 0.001437 Prm2 -0.007012 0.01602 -0.003465 0.001793 -0.001325 Prm3 0.001183 -0.003465 0.0008189 -0.000059 0.0001819 Prm4 -0.007330 0.001793 -0.000059 0.04580 0.003639 Prm5 0.001437 -0.001325 0.0001819 0.003639 0.02029 Prm6 -0.001362 0.0003291 7.7042E-7 -0.002066 -0.002799 Prm7 -0.008647 0.0003538 -0.000031 0.004742 0.007022 Prm8 0.002052 -0.001293 0.0001518 -0.01283 -0.000825 Prm9 -0.000435 0.0004572 0.0000387 -0.000982 -0.005118 Prm10 0.0005566 -0.000569 0.0000644 0.0004195 0.0005391 Covariance Matrix (Empirical) Prm6 Prm7 Prm8 Prm9 Prm10 Prm1 -0.001362 -0.008647 0.002052 -0.000435 0.0005566 Prm2 0.0003291 0.0003538 -0.001293 0.0004572 -0.000569 Prm3 7.7042E-7 -0.000031 0.0001518 0.0000387 0.0000644 Prm4 -0.002066 0.004742 -0.01283 -0.000982 0.0004195 Prm5 -0.002799 0.007022 -0.000825 -0.005118 0.0005391 Prm6 0.01999 -0.002029 0.0004553 0.0006711 -0.005307 Prm7 -0.002029 0.04013 0.0000477 -0.001378 -0.000339 Prm8 0.0004553 0.0000477 0.009373 0.0008017 -0.000426 Prm9 0.0006711 -0.001378 0.0008017 0.004734 -0.000556 Prm10 -0.005307 -0.000339 -0.000426 -0.000556 0.005108 Algorithm converged. Working Correlation Matrix Col1 Col2 Col3 Col4 Row1 1.0000 0.2970 0.3306 0.2552 Row2 0.2970 1.0000 0.4492 0.2839 Row3 0.3306 0.4492 1.0000 0.5104 Row4 0.2552 0.2839 0.5104 1.0000 Analysis Of GEE Parameter Estimates Empirical Standard Error Estimates Standard 95% Confidence Parameter Estimate Error Limits Z Pr > |Z| Intercept -1.0143 0.1165 -1.2427 -0.7860 -8.71 <.0001 time -0.6097 0.1266 -0.8577 -0.3616 -4.82 <.0001 time*time 0.1296 0.0286 0.0736 0.1857 4.53 <.0001 helmert1 0.8110 0.2140 0.3916 1.2305 3.79 0.0002 helmert2 0.3664 0.1424 0.0873 0.6456 2.57 0.0101 helmert3 0.2711 0.1414 -0.0060 0.5482 1.92 0.0552 racew 0.3531 0.2003 -0.0395 0.7457 1.76 0.0779 h1time -0.2194 0.0968 -0.4091 -0.0296 -2.27 0.0235 h2time -0.0729 0.0688 -0.2078 0.0619 -1.06 0.2892 h3time -0.0621 0.0715 -0.2022 0.0780 -0.87 0.3850 Unspecified Working Correlation The GENMOD Procedure .... GEE Model Information .... Working Correlation Matrix Col1 Col2 Col3 Col4 Row1 1.0000 0.3007 0.3324 0.2583 Row2 0.3007 1.0000 0.4513 0.2896 Row3 0.3324 0.4513 1.0000 0.5155 Row4 0.2583 0.2896 0.5155 1.0000 Analysis Of GEE Parameter Estimates Empirical Standard Error Estimates Standard 95% Confidence Parameter Estimate Error Limits Z Pr > |Z| Intercept -1.0102 0.1171 -1.2398 -0.7806 -8.62 <.0001 time -0.6313 0.1310 -0.8880 -0.3746 -4.82 <.0001 time*time 0.1354 0.0297 0.0771 0.1936 4.55 <.0001 helmert1 0.8651 0.2239 0.4263 1.3038 3.86 0.0001 helmert2 0.4310 0.1502 0.1367 0.7254 2.87 0.0041 helmert3 0.2682 0.1481 -0.0221 0.5584 1.81 0.0702 racew 0.3536 0.2001 -0.0386 0.7459 1.77 0.0772 h1time -0.4659 0.2257 -0.9082 -0.0236 -2.06 0.0390 h2time -0.3747 0.1812 -0.7298 -0.0196 -2.07 0.0386 h3time -0.0337 0.1951 -0.4161 0.3487 -0.17 0.8629 time*h1time 0.0624 0.0509 -0.0374 0.1622 1.23 0.2204 time*h2time 0.0786 0.0418 -0.0034 0.1606 1.88 0.0602 time*h3time -0.0070 0.0457 -0.0966 0.0826 -0.15 0.8779 UN Model with interactions - Parameter estimates Obs Parm Estimate Stderr LowerCL UpperCL Z ProbZ 1 Intercept -1.0102 0.1171 -1.2398 -0.7806 -8.62 <.0001 2 time -0.6313 0.1310 -0.8880 -0.3746 -4.82 <.0001 3 time*time 0.1354 0.0297 0.0771 0.1936 4.55 <.0001 4 helmert1 0.8651 0.2239 0.4263 1.3038 3.86 0.0001 5 helmert2 0.4310 0.1502 0.1367 0.7254 2.87 0.0041 6 helmert3 0.2682 0.1481 -0.0221 0.5584 1.81 0.0702 7 racew 0.3536 0.2001 -0.0386 0.7459 1.77 0.0772 8 h1time -0.4659 0.2257 -0.9082 -0.0236 -2.06 0.0390 9 h2time -0.3747 0.1812 -0.7298 -0.0196 -2.07 0.0386 10 h3time -0.0337 0.1951 -0.4161 0.3487 -0.17 0.8629 11 time*h1time 0.0624 0.0509 -0.0374 0.1622 1.23 0.2204 12 time*h2time 0.0786 0.0418 -0.0034 0.1606 1.88 0.0602 13 time*h3time -0.0070 0.0457 -0.0966 0.0826 -0.15 0.8779 UN Model with interactions - Variance-covariances of estimates 12 Row Obs Name Prm1 Prm2 Prm3 Prm4 Prm5 Prm6 Prm7 1 Prm1 0.0137216 -0.007216 0.001224 -0.007952 0.0014305 -0.00142 -0.008522 2 Prm2 -0.007216 0.0171517 -0.003735 0.0034163 -0.001953 0.0006567 -0.000182 3 Prm3 0.001224 -0.003735 0.0008835 -0.00043 0.0003599 -0.000082 0.0001047 4 Prm4 -0.007952 0.0034163 -0.00043 0.0501136 0.0038408 -0.001894 0.0044653 5 Prm5 0.0014305 -0.001953 0.0003599 0.0038408 0.0225562 -0.002588 0.0074913 6 Prm6 -0.00142 0.0006567 -0.000082 -0.001894 -0.002588 0.0219302 -0.001562 7 Prm7 -0.008522 -0.000182 0.0001047 0.0044653 0.0074913 -0.001562 0.0400429 8 Prm8 0.0024934 -0.001314 0.000132 -0.025742 -0.00142 0.000464 0.0019972 9 Prm9 -0.001007 0.0068259 -0.001601 -0.002139 -0.013091 0.0008754 -0.003142 10 Prm10 0.0011545 -0.003354 0.000734 0.0002853 0.000318 -0.012835 -0.002331 11 Prm11 -0.000073 0.0000111 -4.12E-6 0.0031152 0.0001717 -5.231E-6 -0.000482 12 Prm12 0.0001362 -0.001521 0.000386 0.0003161 0.0020034 -0.000057 0.0004688 13 Prm13 -0.000158 0.0006907 -0.000163 0.0000319 0.0000529 0.001935 0.0005031 Obs Prm8 Prm9 Prm10 Prm11 Prm12 Prm13 1 0.0024934 -0.001007 0.0011545 -0.000073 0.0001362 -0.000158 2 -0.001314 0.0068259 -0.003354 0.0000111 -0.001521 0.0006907 3 0.000132 -0.001601 0.000734 -4.12E-6 0.000386 -0.000163 4 -0.025742 -0.002139 0.0002853 0.0031152 0.0003161 0.0000319 5 -0.00142 -0.013091 0.000318 0.0001717 0.0020034 0.0000529 6 0.000464 0.0008754 -0.012835 -5.231E-6 -0.000057 0.001935 7 0.0019972 -0.003142 -0.002331 -0.000482 0.0004688 0.0005031 8 0.0509208 0.006773 -0.003205 -0.010344 -0.001541 0.0006681 9 0.006773 0.0328187 -0.004273 -0.00154 -0.007015 0.0008945 10 -0.003205 -0.004273 0.0380614 0.0006686 0.0008934 -0.008324 11 -0.010344 -0.00154 0.0006686 0.0025942 0.0003899 -0.000157 12 -0.001541 -0.007015 0.0008934 0.0003899 0.0017507 -0.00021 13 0.0006681 0.0008945 -0.008324 -0.000157 -0.00021 0.0020888 Using IML for multi-parameter Wald tests Unspecified GEE model with LIN and QUAD group interactions EXESTM -1.0102 0.1171 -1.2398 -0.7806 -8.6241 0.0000 -0.6313 0.1310 -0.8880 -0.3746 -4.8202 0.0000 0.1354 0.0297 0.0771 0.1936 4.5546 0.0000 0.8651 0.2239 0.4263 1.3038 3.8643 0.0001 0.4310 0.1502 0.1367 0.7254 2.8700 0.0041 0.2682 0.1481 -0.0221 0.5584 1.8109 0.0702 0.3536 0.2001 -0.0386 0.7459 1.7673 0.0772 -0.4659 0.2257 -0.9082 -0.0236 -2.0647 0.0390 -0.3747 0.1812 -0.7298 -0.0196 -2.0683 0.0386 -0.0337 0.1951 -0.4161 0.3487 -0.1727 0.8629 0.0624 0.0509 -0.0374 0.1622 1.2254 0.2204 0.0786 0.0418 -0.0034 0.1606 1.8791 0.0602 -0.0070 0.0457 -0.0966 0.0826 -0.1536 0.8779 EXCOVM COL1 COL2 COL3 COL4 COL5 COL6 COL7 ROW1 0.0137 -0.0072 0.0012 -0.0080 0.0014 -0.0014 -0.0085 ROW2 -0.0072 0.0172 -0.0037 0.0034 -0.0020 0.0007 -0.0002 ROW3 0.0012 -0.0037 0.0009 -0.0004 0.0004 -0.0001 0.0001 ROW4 -0.0080 0.0034 -0.0004 0.0501 0.0038 -0.0019 0.0045 ROW5 0.0014 -0.0020 0.0004 0.0038 0.0226 -0.0026 0.0075 ROW6 -0.0014 0.0007 -0.0001 -0.0019 -0.0026 0.0219 -0.0016 ROW7 -0.0085 -0.0002 0.0001 0.0045 0.0075 -0.0016 0.0400 ROW8 0.0025 -0.0013 0.0001 -0.0257 -0.0014 0.0005 0.0020 ROW9 -0.0010 0.0068 -0.0016 -0.0021 -0.0131 0.0009 -0.0031 ROW10 0.0012 -0.0034 0.0007 0.0003 0.0003 -0.0128 -0.0023 ROW11 -0.0001 0.0000 -0.0000 0.0031 0.0002 -0.0000 -0.0005 ROW12 0.0001 -0.0015 0.0004 0.0003 0.0020 -0.0001 0.0005 ROW13 -0.0002 0.0007 -0.0002 0.0000 0.0001 0.0019 0.0005 EXCOVM COL8 COL9 COL10 COL11 COL12 COL13 ROW1 0.0025 -0.0010 0.0012 -0.0001 0.0001 -0.0002 ROW2 -0.0013 0.0068 -0.0034 0.0000 -0.0015 0.0007 ROW3 0.0001 -0.0016 0.0007 -0.0000 0.0004 -0.0002 ROW4 -0.0257 -0.0021 0.0003 0.0031 0.0003 0.0000 ROW5 -0.0014 -0.0131 0.0003 0.0002 0.0020 0.0001 ROW6 0.0005 0.0009 -0.0128 -0.0000 -0.0001 0.0019 ROW7 0.0020 -0.0031 -0.0023 -0.0005 0.0005 0.0005 ROW8 0.0509 0.0068 -0.0032 -0.0103 -0.0015 0.0007 ROW9 0.0068 0.0328 -0.0043 -0.0015 -0.0070 0.0009 ROW10 -0.0032 -0.0043 0.0381 0.0007 0.0009 -0.0083 ROW11 -0.0103 -0.0015 0.0007 0.0026 0.0004 -0.0002 ROW12 -0.0015 -0.0070 0.0009 0.0004 0.0018 -0.0002 ROW13 0.0007 0.0009 -0.0083 -0.0002 -0.0002 0.0021 EXESTB -1.0102 -0.6313 0.1354 0.8651 0.4310 0.2682 0.3536 -0.4659 -0.3747 -0.0337 0.0624 0.0786 -0.0070 WALD -8.6241 -4.8202 4.5546 3.8643 2.8700 1.8109 1.7673 -2.0647 -2.0683 -0.1727 1.2254 1.8791 -0.1536 Wald statistic - group by QUAD time - 3 df WALD1 4.345 Wald statistic p-value WALD1SIG 0.227 Wald statistic - group by LIN & QUAD time - 6 df WALD1 10.919 Wald statistic p-value WALD1SIG 0.091 Unspecified GEE model with LIN group interactions EXESTM -1.0143 0.1165 -1.2427 -0.7860 -8.7068 0.0000 -0.6097 0.1266 -0.8577 -0.3616 -4.8168 0.0000 0.1296 0.0286 0.0736 0.1857 4.5305 0.0000 0.8110 0.2140 0.3916 1.2305 3.7895 0.0002 0.3664 0.1424 0.0873 0.6456 2.5727 0.0101 0.2711 0.1414 -0.0060 0.5482 1.9173 0.0552 0.3531 0.2003 -0.0395 0.7457 1.7628 0.0779 -0.2194 0.0968 -0.4091 -0.0296 -2.2657 0.0235 -0.0729 0.0688 -0.2078 0.0619 -1.0598 0.2892 -0.0621 0.0715 -0.2022 0.0780 -0.8687 0.3850 EXCOVM COL1 COL2 COL3 COL4 COL5 ROW1 0.0136 -0.0070 0.0012 -0.0073 0.0014 ROW2 -0.0070 0.0160 -0.0035 0.0018 -0.0013 ROW3 0.0012 -0.0035 0.0008 -0.0001 0.0002 ROW4 -0.0073 0.0018 -0.0001 0.0458 0.0036 ROW5 0.0014 -0.0013 0.0002 0.0036 0.0203 ROW6 -0.0014 0.0003 0.0000 -0.0021 -0.0028 ROW7 -0.0086 0.0004 -0.0000 0.0047 0.0070 ROW8 0.0021 -0.0013 0.0002 -0.0128 -0.0008 ROW9 -0.0004 0.0005 0.0000 -0.0010 -0.0051 ROW10 0.0006 -0.0006 0.0001 0.0004 0.0005 EXCOVM COL6 COL7 COL8 COL9 COL10 ROW1 -0.0014 -0.0086 0.0021 -0.0004 0.0006 ROW2 0.0003 0.0004 -0.0013 0.0005 -0.0006 ROW3 0.0000 -0.0000 0.0002 0.0000 0.0001 ROW4 -0.0021 0.0047 -0.0128 -0.0010 0.0004 ROW5 -0.0028 0.0070 -0.0008 -0.0051 0.0005 ROW6 0.0200 -0.0020 0.0005 0.0007 -0.0053 ROW7 -0.0020 0.0401 0.0000 -0.0014 -0.0003 ROW8 0.0005 0.0000 0.0094 0.0008 -0.0004 ROW9 0.0007 -0.0014 0.0008 0.0047 -0.0006 ROW10 -0.0053 -0.0003 -0.0004 -0.0006 0.0051 EXESTB -1.0143 -0.6097 0.1296 0.8110 0.3664 0.2711 0.3531 -0.2194 -0.0729 -0.0621 WALD -8.7068 -4.8168 4.5305 3.7895 2.5727 1.9173 1.7628 -2.2657 -1.0598 -0.8687 Wald statistic - group by LIN time - 3 df WALD1 6.974 Wald statistic p-value WALD1SIG 0.073 H1 estimate at time=4 W1E -0.066 Wald statistic - H1 at time=4 - 1 df WALD1 0.047 Wald statistic p-value WALD1SIG 0.828 H2 estimate at time=4 W1E 0.075 Wald statistic - H2 at time=4 - 1 df WALD1 0.102 Wald statistic p-value WALD1SIG 0.750 H3 estimate at time=4 W1E 0.023 Wald statistic - H3 at time=4 - 1 df WALD1 0.009 Wald statistic p-value WALD1SIG 0.925