- log: /mnt/ide1/home/sschulh1/apc/apc_examplelog log type: text opened on: 21 Jul 2006, 18:08:20 *replicate table 5 and cols 7-9 of table 3 in Yang, Fu and Land (2004) *Stata can maximize GLM objective functions using two different numerical *optimization methods: Newton-Raphson (NR) and iterative reweighted least *squares (IRLS) NR is the default in Stata However, NR presents a *problem in replicating Yang, Fu and Land: The paper set the scale *parameter equal to the deviance divided by the residual degrees of *freedom, but Stata allows this scale parameter only with IRLS and not *with NR So, we show two sets of results below: * 1 NR optimization with scale parameter=pearson chi-squared/residual df * 2 IRLS optimization with scale parameter=deviance/residual df *Version 1 is basically the default in Stata Version 2 matches what was *done in the paper The results are numerically identical to the number of *decimal places shown in the paper use apc_example_datadta *first, using Newton-Raphson optimization (the default in Stata) and scale(x2 > ) * (scale parameter = Pearson chi-squared / residual degrees of freedom) #delim ; delimiter now ; apc_ie death_f if age<=90, > age(age) period(year) cohort(cohort) family(poisson) link(log) > exposure(exp_f) scale(x2); Iteration 0: log likelihood = -8550166 Iteration 1: log likelihood = -40957752 Iteration 2: log likelihood = -10126322 Iteration 3: log likelihood = -97891855 Iteration 4: log likelihood = -97891442 Iteration 5: log likelihood = -97891442 Intrinsic estimator of APC effects No of obs = 152 Optimization : ML Residual df = 102 Scale parameter = 1 Deviance = 1753051016 (1/df) Deviance = 1718677 Pearson = 175173853 (1/df) Pearson = 1717391 AIC = 1294624 Log likelihood = -978914424 BIC = 1701807 OIM age_0 4528543 0158182 2863 0000 4218512 4838574
age_5-2143741 0392401-5463 0000-222065 -2066832 age_10-2353749 0411104-5725 0000-2434324 -2273175 age_15-1704232 0291515-5846 0000-1761367 -1647096 age_20-1629935 0275174-5923 0000-1683869 -1576002 age_25-1570829 0261851-5999 0000-1622151 -1519508 age_30-1377286 0233821-5890 0000-1423115 -1331458 age_35-1091327 0203925-5352 0000-1131295 -1051358 age_40-7506259 0180246-4164 0000-7859534 -7152984 age_45-397673 0156564-2540 0000-428359 -3669869 age_50-0567519 013543-419 0000-0832957 -0302082 age_55 2663666 0117123 2274 0000 2434109 2893223 age_60 6102102 0100392 6078 0000 5905338 6298866 age_65 9559724 008723 10959 0000 9388757 9730691 age_70 1331287 0078762 16903 0000 1315849 1346724 age_75 172417 0075585 22811 0000 1709355 1738984 age_80 2156734 0077788 27726 0000 2141487 217198 age_85 259007 0086002 30116 0000 2573214 2606926 age_90 2988486 0104157 28692 0000 2968072 3008901 period_1960-0389114 0080523-483 0000-0546936 -0231292 period_1965-0090969 0071465-127 0203-0231038 00491 period_1970-0070998 0064584-110 0272-019758 0055585 period_1975-0670092 00616-1088 0000-0790827 -0549357 period_1980-0426419 0059702-714 0000-0543432 -0309405 period_1985 0113779 0060933 187 0062-0005648 0233206 period_1990 0379144 0065691 577 0000 0250392 0507895 period_1995 1154669 0072861 1585 0000 1011864 1297473 cohort_1870 1008335 0311666 3235 0000 9472493 106942 cohort_1875 9767367 0186646 5233 0000 9401547 1013319 cohort_1880 9216318 0138967 6632 0000 8943947 9488688 cohort_1885 8533015 0112908 7557 0000 8311719 8754311 cohort_1890 7756295 0096215 8061 0000 7567718 7944873 cohort_1895 6983224 0085807 8138 0000 6815046 7151402 cohort_1900 6104943 0079727 7657 0000 5948682 6261205 cohort_1905 5215143 007564 6895 0000 5066892 5363394 cohort_1910 455177 0080985 5621 0000 4393042 4710497 cohort_1915 382613 0091723 4171 0000 3646356 4005904 cohort_1920 3169078 0105577 3002 0000 2962151 3376006 cohort_1925 2615428 0122417 2136 0000 2375495 2855361 cohort_1930 1780552 0145238 1226 0000 149589 2065213 cohort_1935 0770235 0172175 447 0000 0432778 1107692 cohort_1940-0668784 0196588-340 0001-1054091 -0283478 cohort_1945-2044153 0214381-954 0000-2464332 -1623974 cohort_1950-2869849 0231115-1242 0000-3322825 -2416872 cohort_1955-3116058 0246563-1264 0000-3599313 -2632804 cohort_1960-3187029 0194917-1635 0000-3569059 -2805 cohort_1965-4604435 0226096-2036 0000-5047576 -4161295 cohort_1970-6198367 0262888-2358 0000-6713619 -5683116 cohort_1975-747627 0303638-2462 0000-8071389 -688115 cohort_1980-9342512 0331409-2819 0000-9992061 -8692963 cohort_1985-1137277 036401-3124 0000-1208622 -1065933 cohort_1990-1342292 0389731-3444 0000-1418677 -1265906 cohort_1995-160697 0480454-3345 0000-1701137 -1512803 _cons -5400347 0060911-88660 0000-5412285 -5388409 apc_cglim death_f if age<=90,
> age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(x2) constraint("a5=a10"); Iteration 0: log likelihood = -8550166 Iteration 1: log likelihood = -40957759 Iteration 2: log likelihood = -1012633 Iteration 3: log likelihood = -97891931 Iteration 4: log likelihood = -97891518 Iteration 5: log likelihood = -97891518 Generalized linear models No of obs = 152 Optimization : ML Residual df = 102 Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740052 (1/df) Pearson = 1717392 AIC = 1294625 Log likelihood = -9789151825 BIC = 1701809 OIM _A_10-2386587 0956633-2495 0000-2574083 -219909 _A_15-152706 1850711-825 0000-1889793 -1164327 _A_20-1242755 2448719-508 0000-1722695 -7628152 _A_25-9736407 3051204-319 0001-1571666 -3756158 _A_30-570089 3654375-156 0119-1286333 1461552 _A_35-0741206 4258852-017 0862-9088402 7605991 _A_40 4765888 4864616 098 0327-4768585 1430036 _A_45 103955 5470944 190 0057-032735 2111836 _A_50 159048 6079696 262 0009 3988814 2782078 _A_55 2123607 668727 318 0001 8129264 3434288 _A_60 2677459 7295039 367 0000 1247658 4107261 _A_65 323323 7902966 409 0000 1684277 4782183 _A_70 3818553 851102 449 0000 2150424 5486682 _A_75 4421445 9119155 485 0000 2634123 6208766 _A_80 5064017 9727349 521 0000 3157492 6970543 _A_85 5707363 1033561 552 0000 3681621 7733104 _A_90 6315787 1094398 577 0000 4170807 8460767 _P_1965-1801941 061602-293 0003-3009319 -0594564 _P_1970-3882057 122038-318 0001-6273958 -1490155 _P_1975-6581237 1827628-360 0000-1016332 -2999152 _P_1980-843765 2435446-346 0001-1321104 -3664264 _P_1985-9997539 3043594-328 0001-1596287 -4032204 _P_1990-1183226 3651418-324 0001-1898891 -4675613 _P_1995-1315682 4260095-309 0002-2150645 -4807189 _C_1875 1784108 0712796 250 0012 0387054 3181162 _C_1880 3333146 1267025 263 0009 0849822 5816469 _C_1885 4749929 1858053 256 0011 1108211 8391647 _C_1890 6073295 2458078 247 0013 1255552 1089104
_C_1895 740031 3061637 242 0016 1399612 1340101 _C_1900 8622116 3666833 235 0019 1435255 1580898 _C_1905 9832402 4272975 230 0021 1457525 1820728 _C_1910 1126911 4880031 231 0021 170443 208338 _C_1915 1264356 5487374 230 0021 1888507 2339862 _C_1920 140866 6095006 231 0021 2140605 2603259 _C_1925 1563303 6702844 233 0020 2495698 2877037 _C_1930 1689824 7310919 231 0021 2569104 3122738 _C_1935 1798801 7919189 227 0023 2466687 3350934 _C_1940 1864908 8527479 219 0029 1935526 3536263 _C_1945 193738 9135616 212 0034 1468317 3727927 _C_1950 2064819 9754629 212 0034 1529464 3976691 _C_1955 2250206 1034978 217 0030 2216865 4278726 _C_1960 2453118 109521 224 0025 3065464 4599689 _C_1965 2521386 1155945 218 0029 2557758 4786996 _C_1970 2572001 1216839 211 0035 1870405 4956962 _C_1975 265422 1277736 208 0038 1499031 5158536 _C_1980 2677604 1338529 200 0045 0541362 5301072 _C_1985 2684587 1399482 192 0055-0583481 5427521 _C_1990 2689581 1458133 184 0065-1683068 5547469 _C_1995 2634911 1521334 173 0083-3468478 561667 _cons -7758225 1094866-709 0000-9904123 -5612326 (Standard errors scaled using square root of Pearson X2-based dispersion) _A_5=_A_10 drop _A* _P* _C*; apc_cglim death_f if age<=90, > age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(x2) constraint("p1965=p1960"); Iteration 0: log likelihood = -8550166 Iteration 1: log likelihood = -40957759 Iteration 2: log likelihood = -1012633 Iteration 3: log likelihood = -97891931 Iteration 4: log likelihood = -97891518 Iteration 5: log likelihood = -97891518 Generalized linear models No of obs = 152 Optimization : ML Residual df = 102 Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740052 (1/df) Pearson = 1717392 AIC = 1294625 Log likelihood = -9789151825 BIC = 1701809 OIM
_A_5-2566781 0451768-5682 0000-2655326 -2478236 _A_10-2746975 0501999-5472 0000-2845365 -2648585 _A_15-2067642 0450515-4590 0000-2155942 -1979343 _A_20-1963532 0511821-3836 0000-2063847 -1863217 _A_25-1874611 0586489-3196 0000-1989561 -1759662 _A_30-1651254 0663025-2490 0000-1781204 -1521304 _A_35-133548 0742786-1798 0000-1481063 -1189896 _A_40-9649644 0835718-1155 0000-1128762 -8011666 _A_45-5821971 0924043-630 0000-7633061 -401088 _A_50-2114616 1014431-208 0037-4102863 -0126368 _A_55 1414715 1106156 128 0201-0753312 3582741 _A_60 5151295 1199278 430 0000 2800754 7501836 _A_65 8907062 1293084 689 0000 6372663 1144146 _A_70 1295835 138728 934 0000 1023933 1567737 _A_75 1718533 1481719 1160 0000 1428121 2008944 _A_80 2180911 1576125 1384 0000 1871996 2489826 _A_85 2644062 1670493 1583 0000 2316651 2971473 _A_90 3072293 1763944 1742 0000 2726566 3418019 _P_1970-0278174 0163316-170 0089-0598267 004192 _P_1975-1175413 0252372-466 0000-1670053 -0680772 _P_1980-1229884 0345423-356 0000-1906901 -0552866 _P_1985-0987831 0440429-224 0025-1851056 -0124606 _P_1990-1020611 0536473-190 0057-2072079 0030857 _P_1995-0543231 0632841-086 0391-1783577 0697116 _C_1875-0017833 0374126-005 0962-0751106 0715439 _C_1880-0270737 0378411-072 0474-1012409 0470934 _C_1885-0655896 0416626-157 0115-1472467 0160675 _C_1890-1134471 0475724-238 0017-2066873 -0202068 _C_1895-1609397 0547136-294 0003-2681763 -0537031 _C_1900-2189533 0626267-350 0000-3416993 -0962073 _C_1905-2781189 071001-392 0000-4172783 -1389594 _C_1910-3146417 0797978-394 0000-4710425 -1582409 _C_1915-3573912 0887308-403 0000-5313003 -183482 _C_1920-3932819 0978511-402 0000-5850665 -2014973 _C_1925-4188324 1071498-391 0000-6288422 -2088226 _C_1930-4725056 1166313-405 0000-7010988 -2439125 _C_1935-5437228 1262841-431 0000-791235 -2962106 _C_1940-6578103 1360103-484 0000-9243855 -3912351 _C_1945-7655327 1457047-525 0000-1051109 -4799566 _C_1950-8182878 1553349-527 0000-1122739 -513837 _C_1955-8130943 1653866-492 0000-1137246 -4889425 _C_1960-7903768 1767803-447 0000-113686 -4438939 _C_1965-902303 1818861-496 0000-1258793 -5458127 _C_1970-1031882 1945369-530 0000-1413167 -6505964 _C_1975-1129858 204748-552 0000-1531156 -7285588 _C_1980-1286667 2147892-599 0000-1707646 -8656882 _C_1985-1459879 2249617-649 0000-1900796 -1018962 _C_1990-1635079 2350308-696 0000-2095731 -1174427 _C_1995-1869943 2456334-761 0000-2351375 -138851 _cons -451473 1792801-2518 0000-4866112 -4163347 (Standard errors scaled using square root of Pearson X2-based dispersion) _P_1965=_P_1960
drop _A* _P* _C*; apc_cglim death_f if age<=90, > age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(x2) constraint("c1995=c1990"); Iteration 0: log likelihood = -8550166 Iteration 1: log likelihood = -40957759 Iteration 2: log likelihood = -1012633 Iteration 3: log likelihood = -97891931 Iteration 4: log likelihood = -97891518 Iteration 5: log likelihood = -97891518 Generalized linear models No of obs = 152 Optimization : ML Residual df = 102 Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740054 (1/df) Pearson = 1717392 AIC = 1294625 Log likelihood = -9789151825 BIC = 1701809 OIM _A_5-2331917 0758645-3074 0000-2480608 -2183225 _A_10-2277247 1360273-1674 0000-2543856 -2010638 _A_15-1363051 1949943-699 0000-1745232 -9808689 _A_20-1024076 2581374-397 0000-1530016 -5181363 _A_25-7002919 3216904-218 0029-1330793 -0697903 _A_30-2420704 3853203-063 0530-9972844 5131435 _A_35 3085678 4490753 069 0492-5716035 1188739 _A_40 9139469 5129686 178 0075-091453 1919347 _A_45 1531578 5768638 266 0008 4009459 2662211 _A_50 2137178 6407812 334 0001 8812695 3393086 _A_55 2724975 7047254 387 0000 1343738 4106211 _A_60 3333497 7686814 434 0000 1826909 4840084 _A_65 3943937 8326568 474 0000 231196 5575914 _A_70 458393 8966368 511 0000 2826554 6341306 _A_75 5241491 9606305 546 0000 335869 7124293 _A_80 5938734 1024632 580 0000 3930492 7946975 _A_85 6636749 1088639 610 0000 4503056 8770441 _A_90 7299843 1152656 633 0000 5040679 9559007 _P_1965-2348639 0647705-363 0000-3618117 -1079162 _P_1970-4975452 128453-387 0000-7493084 -245782 _P_1975-822133 192376-427 0000-1199183 -4450829 _P_1980-1062444 2563479-414 0000-1564877 -5600115 _P_1985-1273103 3203428-397 0000-1900963 -6452424 _P_1990-1511245 3848444-393 0000-2265526 -7569634 _P_1995-1698371 4478893-379 0000-2576217 -8205236
_C_1875 2330806 0740208 315 0002 0880025 3781587 _C_1880 4426541 132847 333 0001 1822788 7030294 _C_1885 6390022 1952258 327 0001 2563666 1021638 _C_1890 8260086 2584628 320 0001 3194309 1332586 _C_1895 101338 3220324 315 0002 3822081 1644552 _C_1900 119023 3858337 308 0002 4340101 194645 _C_1905 1365929 4495961 304 0002 4847364 2247121 _C_1910 156427 5134668 305 0002 5578932 2570646 _C_1915 1756384 577363 304 0002 6247734 2887995 _C_1920 1955357 6412955 305 0002 6984412 3212273 _C_1925 2164671 7052564 307 0002 7823936 3546948 _C_1930 2345861 7692407 305 0002 8381772 3853546 _C_1935 2509508 8332518 301 0003 8763647 4142652 _C_1940 2630285 8972739 293 0003 87166 4388909 _C_1945 2757426 9612924 287 0004 8733276 4641525 _C_1950 2939535 1025325 287 0004 9299345 4949135 _C_1955 3179592 1089485 292 0004 1044241 5314943 _C_1960 3437174 1153298 298 0003 1176752 5697596 _C_1965 3560111 1217397 292 0003 1174058 5946165 _C_1970 3665397 1281497 286 0004 1153709 6177084 _C_1975 3802285 1345617 283 0005 1164923 6439646 _C_1980 3880339 1409686 275 0006 1117404 6643274 _C_1985 3941991 1473819 267 0007 105336 6830623 _C_1990 4001655 1565265 256 0011 9337915 7069519 _cons -8742281 1153101-758 0000-1100232 -6482244 (Standard errors scaled using square root of Pearson X2-based dispersion) _C_1995=_C_1990 drop _A* _P* _C*; #delim cr delimiter now cr *next, using IRLS optimization (the default in S-Plus) and scale(dev) * (scale parameter = deviance / residual degrees of freedom) #delim ; delimiter now ; apc_ie death_f if age<=90, > age(age) period(year) cohort(cohort) family(poisson) link(log) > exposure(exp_f) scale(dev) irls; Iteration 1: deviance = 1707985 Iteration 2: deviance = 1970306 Iteration 3: deviance = 2651635 Iteration 4: deviance = 1759935 Iteration 5: deviance = 1753052 Iteration 6: deviance = 1753051 Iteration 7: deviance = 1753051 Intrinsic estimator of APC effects No of obs = 152 Optimization : MQL Fisher scoring Residual df = 102 (IRLS EIM) Scale parameter = 1 Deviance = 1753051016 (1/df) Deviance = 1718677 Pearson = 1751738529 (1/df) Pearson = 1717391
AIC = Deviance = BIC = 1701807 EIM age_0 4528543 0158241 2862 0000 4218396 483869 age_5-2143741 0392548-5461 0000-2220679 -2066803 age_10-2353749 0411258-5723 0000-2434354 -2273144 age_15-1704232 0291624-5844 0000-1761389 -1647074 age_20-1629935 0275277-5921 0000-1683889 -1575982 age_25-1570829 0261949-5997 0000-1622171 -1519488 age_30-1377286 0233909-5888 0000-1423132 -1331441 age_35-1091327 0204001-5350 0000-113131 -1051343 age_40-7506259 0180313-4163 0000-7859666 -7152851 age_45-397673 0156623-2539 0000-4283705 -3669754 age_50-0567519 013548-419 0000-0833056 -0301983 age_55 2663666 0117167 2273 0000 2434023 2893309 age_60 6102102 0100429 6076 0000 5905264 629894 age_65 9559724 0087262 10955 0000 9388693 9730755 age_70 1331287 0078792 16896 0000 1315844 1346729 age_75 172417 0075614 22802 0000 170935 173899 age_80 2156734 0077818 27715 0000 2141482 2171986 age_85 259007 0086034 30105 0000 2573208 2606933 age_90 2988486 0104196 28681 0000 2968064 3008909 period_1960-0389114 0080553-483 0000-0546995 -0231233 period_1965-0090969 0071492-127 0203-023109 0049153 period_1970-0070998 0064608-110 0272-0197628 0055632 period_1975-0670092 0061624-1087 0000-0790872 -0549312 period_1980-0426419 0059724-714 0000-0543476 -0309362 period_1985 0113779 0060956 187 0062-0005693 0233251 period_1990 0379144 0065715 577 0000 0250344 0507944 period_1995 1154669 0072888 1584 0000 1011811 1297527 cohort_1870 1008335 0311782 3234 0000 9472264 1069443 cohort_1875 9767367 0186716 5231 0000 940141 1013332 cohort_1880 9216318 0139019 6630 0000 8943845 948879 cohort_1885 8533015 0112951 7555 0000 8311636 8754394 cohort_1890 7756295 0096251 8058 0000 7567647 7944944 cohort_1895 6983224 0085839 8135 0000 6814983 7151465 cohort_1900 6104943 0079756 7654 0000 5948624 6261263 cohort_1905 5215143 0075668 6892 0000 5066836 5363449 cohort_1910 455177 0081015 5618 0000 4392983 4710557 cohort_1915 382613 0091757 4170 0000 3646289 4005972 cohort_1920 3169078 0105617 3001 0000 2962073 3376083 cohort_1925 2615428 0122463 2136 0000 2375405 2855451 cohort_1930 1780552 0145292 1225 0000 1495784 206532 cohort_1935 0770235 017224 447 0000 0432651 1107819 cohort_1940-0668784 0196662-340 0001-1054235 -0283334 cohort_1945-2044153 0214461-953 0000-246449 -1623816 cohort_1950-2869849 0231201-1241 0000-3322995 -2416702 cohort_1955-3116058 0246655-1263 0000-3599494 -2632623 cohort_1960-3187029 019499-1634 0000-3569202 -2804857 cohort_1965-4604435 0226181-2036 0000-5047742 -4161129
cohort_1970-6198367 0262987-2357 0000-6713812 -5682923 cohort_1975-747627 0303752-2461 0000-8071612 -6880927 cohort_1980-9342512 0331533-2818 0000-9992304 -869272 cohort_1985-1137277 0364146-3123 0000-1208648 -1065906 cohort_1990-1342292 0389877-3443 0000-1418706 -1265877 cohort_1995-160697 0480634-3343 0000-1701172 -1512767 _cons -5400347 0060933-88627 0000-541229 -5388404 apc_cglim death_f if age<=90, > age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(dev) irls constraint("a5=a10"); Iteration 1: deviance = 1707985 Iteration 2: deviance = 1970306 Iteration 3: deviance = 2651637 Iteration 4: deviance = 1759937 Iteration 5: deviance = 1753053 Iteration 6: deviance = 1753053 Iteration 7: deviance = 1753053 Generalized linear models No of obs = 152 Optimization : MQL Fisher scoring Residual df = 102 (IRLS EIM) Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740051 (1/df) Pearson = 1717392 BIC = 1701809 EIM _A_10-2386587 0956991-2494 0000-2574153 -219902 _A_15-152706 1851404-825 0000-1889928 -1164192 _A_20-1242755 2449636-507 0000-1722875 -7626354 _A_25-9736407 3052347-319 0001-157189 -3753918 _A_30-570089 3655743-156 0119-1286602 1464235 _A_35-0741206 4260447-017 0862-9091528 7609117 _A_40 4765888 4866438 098 0327-4772156 1430393 _A_45 103955 5472993 190 0058-0331366 2112237 _A_50 159048 6081973 262 0009 3984351 2782525 _A_55 2123607 6689774 317 0002 8124354 3434779 _A_60 2677459 7297771 367 0000 1247122 4107796 _A_65 323323 7905926 409 0000 1683697 4782763 _A_70 3818553 8514207 448 0000 2149799 5487307 _A_75 4421445 912257 485 0000 2633454 6209436 _A_80 5064017 9730992 520 0000 3156778 6971257 _A_85 5707363 1033948 552 0000 3680862 7733863 _A_90 6315787 1094808 577 0000 4170004 8461571
_P_1965-1801941 0616251-292 0003-3009771 -0594112 _P_1970-3882057 1220838-318 0001-6274854 -1489259 _P_1975-6581237 1828313-360 0000-1016466 -299781 _P_1980-843765 2436358-346 0001-1321282 -3662476 _P_1985-9997539 3044734-328 0001-1596511 -4029969 _P_1990-1183226 3652785-324 0001-1899159 -4672933 _P_1995-1315682 426169-309 0002-2150958 -4804061 _C_1875 1784108 0713063 250 0012 0386531 3181686 _C_1880 3333146 12675 263 0009 0848891 58174 _C_1885 4749929 1858749 256 0011 1106847 8393011 _C_1890 6073295 2458998 247 0014 1253747 1089284 _C_1895 740031 3062784 242 0016 1397364 1340326 _C_1900 8622116 3668207 235 0019 1432563 1581167 _C_1905 9832402 4274575 230 0021 1454389 1821041 _C_1910 1126911 4881859 231 0021 1700848 2083738 _C_1915 1264356 5489429 230 0021 1884479 2340264 _C_1920 140866 6097288 231 0021 213613 2603706 _C_1925 1563303 6705355 233 0020 2490778 2877529 _C_1930 1689824 7313657 231 0021 2563737 3123275 _C_1935 1798801 7922155 227 0023 2460873 3351515 _C_1940 1864908 8530673 219 0029 1929266 3536889 _C_1945 193738 9139038 212 0034 1461611 3728598 _C_1950 2064819 9758283 212 0034 1522304 3977407 _C_1955 2250206 1035366 217 0030 2209268 4279486 _C_1960 2453118 109562 224 0025 3057424 4600493 _C_1965 2521386 1156378 218 0029 2549272 4787845 _C_1970 2572001 1217295 211 0035 1861473 4957855 _C_1975 265422 1278215 208 0038 1489652 5159474 _C_1980 2677604 133903 200 0046 0531536 5302054 _C_1985 2684587 1400006 192 0055-0593755 5428549 _C_1990 2689581 1458679 184 0065-1693772 5548539 _C_1995 2634911 1521903 173 0083-3479645 5617787 _cons -7758225 1095277-708 0000-9904927 -5611522 (Standard errors scaled using square root of deviance-based dispersion) _A_5=_A_10 drop _A* _P* _C*; apc_cglim death_f if age<=90, > age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(dev) irls constraint("p1965=p1960"); Iteration 1: deviance = 1707985 Iteration 2: deviance = 1970306 Iteration 3: deviance = 2651637 Iteration 4: deviance = 1759937 Iteration 5: deviance = 1753053 Iteration 6: deviance = 1753053 Iteration 7: deviance = 1753053 Generalized linear models No of obs = 152 Optimization : MQL Fisher scoring Residual df = 102
(IRLS EIM) Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740051 (1/df) Pearson = 1717392 BIC = 1701809 EIM _A_5-2566781 0451937-5680 0000-2655359 -2478203 _A_10-2746975 0502187-5470 0000-2845402 -2648548 _A_15-2067642 0450684-4588 0000-2155975 -197931 _A_20-1963532 0512012-3835 0000-2063884 -1863179 _A_25-1874611 0586709-3195 0000-1989604 -1759619 _A_30-1651254 0663273-2490 0000-1781253 -1521255 _A_35-133548 0743064-1797 0000-1481117 -1189842 _A_40-9649644 0836031-1154 0000-1128824 -8011053 _A_45-5821971 0924389-630 0000-7633739 -4010202 _A_50-2114616 1014811-208 0037-4103608 -0125624 _A_55 1414715 1106571 128 0201-0754124 3583553 _A_60 5151295 1199727 429 0000 2799874 7502717 _A_65 8907062 1293569 689 0000 6371714 1144241 _A_70 1295835 13878 934 0000 1023831 1567839 _A_75 1718533 1482274 1159 0000 1428012 2009053 _A_80 2180911 1576715 1383 0000 187188 2489941 _A_85 2644062 1671118 1582 0000 2316529 2971595 _A_90 3072293 1764605 1741 0000 2726437 3418149 _P_1970-0278174 0163377-170 0089-0598387 0042039 _P_1975-1175413 0252467-466 0000-1670239 -0680587 _P_1980-1229884 0345553-356 0000-1907155 -0552613 _P_1985-0987831 0440594-224 0025-185138 -0124283 _P_1990-1020611 0536674-190 0057-2072473 003125 _P_1995-0543231 0633078-086 0391-1784042 069758 _C_1875-0017833 0374266-005 0962-075138 0715714 _C_1880-0270737 0378552-072 0474-1012686 0471212 _C_1885-0655896 0416782-157 0116-1472773 0160981 _C_1890-1134471 0475902-238 0017-2067222 -0201719 _C_1895-1609397 054734-294 0003-2682165 -0536629 _C_1900-2189533 0626501-349 0000-3417453 -0961613 _C_1905-2781189 0710276-392 0000-4173305 -1389073 _C_1910-3146417 0798277-394 0000-471101 -1581824 _C_1915-3573912 088764-403 0000-5313655 -1834168 _C_1920-3932819 0978877-402 0000-5851383 -2014254 _C_1925-4188324 10719-391 0000-6289209 -208744 _C_1930-4725056 116675-405 0000-7011844 -2438268 _C_1935-5437228 1263314-430 0000-7913277 -2961179 _C_1940-6578103 1360612-483 0000-9244854 -3911352 _C_1945-7655327 1457593-525 0000-1051216 -4798497 _C_1950-8182878 1553931-527 0000-1122853 -5137229 _C_1955-8130943 1654485-491 0000-1137367 -4888211 _C_1960-7903768 1768465-447 0000-113699 -4437641 _C_1965-902303 1819542-496 0000-1258927 -5456792 _C_1970-1031882 1946098-530 0000-141331 -6504536
_C_1975-1129858 2048247-552 0000-1531307 -7284084 _C_1980-1286667 2148696-599 0000-1707804 -8655305 _C_1985-1459879 225046-649 0000-1900961 -1018797 _C_1990-1635079 2351188-695 0000-2095903 -1174254 _C_1995-1869943 2457254-761 0000-2351556 -138833 _cons -451473 1793472-2517 0000-4866244 -4163216 (Standard errors scaled using square root of deviance-based dispersion) _P_1965=_P_1960 drop _A* _P* _C*; apc_cglim death_f if age<=90, > age(age) period(year) cohort(cohort) > agepfx("_a") periodpfx("_p") cohortpfx("_c") > family(poisson) link(log) > exposure(exp_f) scale(dev) irls constraint("c1995=c1990"); Iteration 1: deviance = 1707985 Iteration 2: deviance = 1970306 Iteration 3: deviance = 2651637 Iteration 4: deviance = 1759937 Iteration 5: deviance = 1753053 Iteration 6: deviance = 1753053 Iteration 7: deviance = 1753053 Generalized linear models No of obs = 152 Optimization : MQL Fisher scoring Residual df = 102 (IRLS EIM) Scale parameter = 1 Deviance = 1753052533 (1/df) Deviance = 1718679 Pearson = 1751740051 (1/df) Pearson = 1717392 BIC = 1701809 EIM _A_5-2331917 0758929-3073 0000-2480664 -2183169 _A_10-2277247 1360783-1673 0000-2543955 -2010538 _A_15-1363051 1950673-699 0000-1745376 -9807258 _A_20-1024076 258234-397 0000-1530206 -5179468 _A_25-7002919 3218109-218 0030-133103 -0695542 _A_30-2420704 3854646-063 0530-9975672 5134264 _A_35 3085678 4492435 069 0492-5719332 1189069 _A_40 9139469 5131607 178 0075-0918296 1919723 _A_45 1531578 5770799 265 0008 4005224 2662634 _A_50 2137178 6410212 333 0001 8807991 3393556 _A_55 2724975 7049894 387 0000 1343221 4106728 _A_60 3333497 7689693 434 0000 1826344 4840649 _A_65 3943937 8329687 473 0000 2311349 5576526 _A_70 458393 8969726 511 0000 2825896 6341964
_A_75 5241491 9609903 545 0000 3357985 7124998 _A_80 5938734 1025016 579 0000 392974 7947727 _A_85 6636749 1089047 609 0000 4502257 8771241 _A_90 7299843 1153088 633 0000 5039833 9559853 _P_1965-2348639 0647947-362 0000-3618592 -1078686 _P_1970-4975452 1285011-387 0000-7494027 -2456877 _P_1975-822133 1924481-427 0000-1199324 -4449417 _P_1980-1062444 2564439-414 0000-1565065 -5598233 _P_1985-1273103 3204628-397 0000-1901198 -6450072 _P_1990-1511245 3849886-393 0000-2265808 -7566809 _P_1995-1698371 448057-379 0000-2576546 -8201948 _C_1875 2330806 0740485 315 0002 0879482 378213 _C_1880 4426541 1328968 333 0001 1821812 7031269 _C_1885 6390022 1952989 327 0001 2562233 1021781 _C_1890 8260086 2585596 319 0001 3192412 1332776 _C_1895 101338 322153 315 0002 3819717 1644788 _C_1900 119023 3859782 308 0002 4337269 1946734 _C_1905 1365929 4497645 304 0002 4844063 2247451 _C_1910 156427 5136592 305 0002 5575162 2571023 _C_1915 1756384 5775793 304 0002 6243495 2888419 _C_1920 1955357 6415357 305 0002 6979704 3212744 _C_1925 2164671 7055205 307 0002 7818758 3547466 _C_1930 2345861 7695289 305 0002 8376125 385411 _C_1935 2509508 8335639 301 0003 875753 4143263 _C_1940 2630285 89761 293 0003 8710013 4389568 _C_1945 2757426 9616525 287 0004 8726219 464223 _C_1950 2939535 1025709 287 0004 9291818 4949888 _C_1955 3179592 1089893 292 0004 1043442 5315743 _C_1960 3437174 115373 298 0003 1175905 5698442 _C_1965 3560111 1217853 292 0003 1173164 5947059 _C_1970 3665397 1281977 286 0004 1152768 6178025 _C_1975 3802285 1346121 282 0005 1163935 6440634 _C_1980 3880339 1410214 275 0006 111637 6644308 _C_1985 3941991 1474371 267 0008 1052278 6831705 _C_1990 4001655 1565852 256 0011 9326424 7070668 _cons -8742281 1153533-758 0000-1100316 -6481398 (Standard errors scaled using square root of deviance-based dispersion) _C_1995=_C_1990 drop _A* _P* _C*; #delim cr delimiter now cr log close log: /mnt/ide1/home/sschulh1/apc/apc_examplelog log type: text closed on: 21 Jul 2006, 18:08:28 -