--> RESET --> Read ; Nobs = 62 ; Nvar = 2 ; Names = 1 $ --> Create ; LogT = Log(T) $ --> SURVIVAL ; lhs = LogT ; plot $ +------------------------------------------------+ | Estimated Survival Function | | Duration variable is LOGT | | Status is given by variable ONE | +------------------------------------------------+ Number of observations in stratum = 62 Number of observations exiting = 62 Number of observations censored = 0 Survival Enter Cnsrd At Risk Exited Survival Rate Hazard Rate .0- .5 62 0 62 1 1.0000 ( .000) .0302 ( .030) .5- 1.1 61 0 61 3 .9839 ( .016) .0938 ( .054) 1.1- 1.6 58 0 58 7 .9355 ( .031) .2389 ( .090) 1.6- 2.2 51 0 51 2 .8226 ( .049) .0744 ( .053) 2.2- 2.7 49 0 49 8 .7903 ( .052) .3307 ( .116) 2.7- 3.2 41 0 41 8 .6613 ( .060) .4022 ( .141) 3.2- 3.8 33 0 33 14 .5323 ( .063) 1.0017 ( .258) 3.8- 4.3 19 0 19 7 .3065 ( .059) .8402 ( .309) 4.3- 4.8 12 0 12 8 .1935 ( .050) 1.8604 ( .570) 4.8- 5.4 4 0 4 4 .0645 ( .031) 3.7207 ( .000) --> SURVIVAL ; lhs = T ; rhs = Prod $ +---------------------------------------------------+ | Cox Proportional Hazard Model | | Duration variable is T | | Status is given by variable ONE | | Total Number of Observations = 62 | | Total Number of Observations Exiting = 62 | | Total Number of Observations Censored = 0 | | Total Number of Distinct Exit Times = 49 | | Number of Observed Times Incl. Cnsrd. = 49 | +---------------------------------------------------+ Normal exit from iterations. Exit status=0. +---------------------------------------------+ | Cox Proportional Hazard Model | | Maximum Likelihood Estimates | | Model estimated: Nov 20, 2004 at 02:36:23AM.| | Dependent variable T | | Weighting variable None | | Number of observations 62 | | Iterations completed 4 | | Log likelihood function -193.2722 | | Restricted log likelihood -197.3651 | | Chi squared 8.185853 | | Degrees of freedom 1 | | Prob[ChiSqd > value] = .4221833E-02 | | Log-rank test with 1 degrees of freedom: | | Chi-squared = 8.064, Prob = .0045 | +---------------------------------------------+ +---------+--------------+----------------+--------+---------+----------+ |Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X| +---------+--------------+----------------+--------+---------+----------+ PROD 9.073034850 3.2252989 2.813 .0049 .11023065E-01 --> SURVIVAL ; lhs = LogT ; Model = Weibull ; rhs = One, Prod ; plot ; list $ Normal exit from iterations. Exit status=0. +---------------------------------------------+ | Loglinear survival model: WEIBULL | | Maximum Likelihood Estimates | | Model estimated: Nov 20, 2004 at 02:36:23AM.| | Dependent variable LOGT | | Weighting variable None | | Number of observations 62 | | Iterations completed 7 | | Log likelihood function -97.28542 | +---------------------------------------------+ +---------+--------------+----------------+--------+---------+----------+ |Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X| +---------+--------------+----------------+--------+---------+----------+ RHS of hazard model Constant 3.779774234 .13833256 27.324 .0000 PROD -9.332198509 2.9542845 -3.159 .0016 .11023065E-01 Ancillary parameters for survival Sigma .9922036633 .12063878 8.225 .0000 +----------------------------------------------------------------+ | Parameters of underlying density at data means: | | Parameter Estimate Std. Error Confidence Interval | | ------------------------------------------------------------ | | Lambda .02530 .00337 .0187 to .0319 | | P 1.00786 .12254 .7677 to 1.2480 | | Median 27.47425 3.66307 20.2946 to 34.6539 | | Percentiles of survival distribution: | | Survival .25 .50 .75 .95 | | Time 54.65 27.47 11.48 2.07 | +----------------------------------------------------------------+ Predicted Values (* => observation was not in estimating sample.) Observation Obsrvd.Time exp(bx)=prdT Intg.Hazard Hazard Survival 1 7.0000 39.392 .1753 .0252 .8392 2 14.000 39.392 .3525 .0254 .7029 3 52.000 39.392 1.3229 .0256 .2664 4 37.000 35.347 1.0471 .0285 .3509 5 52.000 35.347 1.4756 .0286 .2286 6 17.000 63.374 .2655 .0157 .7668 7 72.000 63.374 1.1373 .0159 .3207 8 114.00 63.374 1.8072 .0160 .1641 9 216.00 63.374 3.4414 .0161 .0320 10 98.000 72.964 1.3462 .0138 .2602 11 85.000 41.673 2.0512 .0243 .1286 12 1.0000 23.995 .0406 .0410 .9602 13 3.0000 23.995 .1230 .0413 .8843 14 8.0000 23.995 .3305 .0416 .7185 15 23.000 23.995 .9582 .0420 .3836 16 33.000 23.995 1.3787 .0421 .2519 17 43.000 23.995 1.8003 .0422 .1653 18 5.0000 116.09 .0420 .0085 .9589 19 12.000 46.763 .2539 .0213 .7758 20 21.000 46.763 .4463 .0214 .6400 21 42.000 46.763 .8974 .0215 .4076 22 9.0000 39.392 .2258 .0253 .7978 23 26.000 39.392 .6579 .0255 .5180 24 130.00 39.392 3.3312 .0258 .0357 25 41.000 35.347 1.1613 .0285 .3131 26 119.00 35.347 3.3988 .0288 .0334 27 19.000 63.374 .2970 .0158 .7431 28 99.000 63.374 1.5676 .0160 .2085 29 152.00 63.374 2.4150 .0160 .0894 30 15.000 72.964 .2030 .0136 .8162 31 2.0000 41.673 .0469 .0236 .9542 32 3.0000 21.904 .1348 .0453 .8739 33 2.0000 23.995 .0817 .0412 .9215 34 3.0000 23.995 .1230 .0413 .8843 35 11.000 23.995 .4556 .0417 .6341 36 27.000 23.995 1.1263 .0420 .3242 37 35.000 23.995 1.4630 .0421 .2315 38 44.000 23.995 1.8425 .0422 .1584 39 49.000 116.09 .4192 .0086 .6575 40 12.000 46.763 .2539 .0213 .7758 41 27.000 46.763 .5749 .0215 .5628 42 117.00 46.763 2.5201 .0217 .0805 43 13.000 39.392 .3272 .0254 .7210 44 29.000 39.392 .7344 .0255 .4798 45 9.0000 35.347 .2519 .0282 .7773 46 49.000 35.347 1.3898 .0286 .2491 47 3.0000 63.374 .0462 .0155 .9548 48 28.000 63.374 .4390 .0158 .6447 49 104.00 63.374 1.6475 .0160 .1925 50 153.00 63.374 2.4310 .0160 .0879 51 61.000 72.964 .8349 .0138 .4339 52 25.000 41.673 .5975 .0241 .5502 53 10.000 21.904 .4537 .0457 .6353 54 3.0000 23.995 .1230 .0413 .8843 55 4.0000 23.995 .1644 .0414 .8484 56 22.000 23.995 .9162 .0420 .4000 57 32.000 23.995 1.3366 .0421 .2627 58 43.000 23.995 1.8003 .0422 .1653 59 100.00 23.995 4.2145 .0425 .0148 60 2.0000 46.763 .0417 .0210 .9591 61 21.000 46.763 .4463 .0214 .6400 62 38.000 46.763 .8113 .0215 .4443 --> SURVIVAL ; lhs = LogT ; Model=Exponential ; rhs = One, Prod ? ; plot ; list $ Note: DFP and BFGS usually take more than 4 or 5 iterations to converge. If this problem was not structured for quick convergence, you might want to examine results closely. If convergence is too early, tighten convergence with, e.g., ;TLG=1.D-9. Normal exit from iterations. Exit status=0. +---------------------------------------------+ | Loglinear survival model: EXPONENTIAL | | Maximum Likelihood Estimates | | Model estimated: Nov 20, 2004 at 02:36:23AM.| | Dependent variable LOGT | | Weighting variable None | | Number of observations 62 | | Iterations completed 4 | | Log likelihood function -97.28844 | +---------------------------------------------+ +---------+--------------+----------------+--------+---------+----------+ |Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X| +---------+--------------+----------------+--------+---------+----------+ RHS of hazard model Constant 3.776511939 .13908855 27.152 .0000 PROD -9.333813813 2.9778671 -3.134 .0017 .11023065E-01 Ancillary parameters for survival Sigma 1.000000000 ........(Fixed Parameter)........ +----------------------------------------------------------------+ | Parameters of underlying density at data means: | | Parameter Estimate Std. Error Confidence Interval | | ------------------------------------------------------------ | | Lambda .02538 .00339 .0187 to .0320 | | P 1.00000 .00000 1.0000 to 1.0000 | | Median 27.30615 3.64417 20.1636 to 34.4487 | | Percentiles of survival distribution: | | Survival .25 .50 .75 .95 | | Time 54.61 27.31 11.33 2.02 | +----------------------------------------------------------------+ Predicted Values (* => observation was not in estimating sample.) Observation Obsrvd.Time exp(bx)=prdT Intg.Hazard Hazard Survival 1 7.0000 39.263 .1783 .0255 .8367 2 14.000 39.263 .3566 .0255 .7001 3 52.000 39.263 1.3244 .0255 .2660 4 37.000 35.231 1.0502 .0284 .3499 5 52.000 35.231 1.4760 .0284 .2286 6 17.000 63.171 .2691 .0158 .7641 7 72.000 63.171 1.1398 .0158 .3199 8 114.00 63.171 1.8046 .0158 .1645 9 216.00 63.171 3.4193 .0158 .0327 10 98.000 72.733 1.3474 .0137 .2599 11 85.000 41.537 2.0464 .0241 .1292 12 1.0000 23.914 .0418 .0418 .9590 13 3.0000 23.914 .1254 .0418 .8821 14 8.0000 23.914 .3345 .0418 .7157 15 23.000 23.914 .9618 .0418 .3822 16 33.000 23.914 1.3799 .0418 .2516 17 43.000 23.914 1.7981 .0418 .1656 18 5.0000 115.73 .0432 .0086 .9577 19 12.000 46.612 .2574 .0215 .7730 20 21.000 46.612 .4505 .0215 .6373 21 42.000 46.612 .9011 .0215 .4061 22 9.0000 39.263 .2292 .0255 .7952 23 26.000 39.263 .6622 .0255 .5157 24 130.00 39.263 3.3110 .0255 .0365 25 41.000 35.231 1.1637 .0284 .3123 26 119.00 35.231 3.3777 .0284 .0341 27 19.000 63.171 .3008 .0158 .7402 28 99.000 63.171 1.5672 .0158 .2086 29 152.00 63.171 2.4062 .0158 .0902 30 15.000 72.733 .2062 .0137 .8136 31 2.0000 41.537 .0482 .0241 .9530 32 3.0000 21.830 .1374 .0458 .8716 33 2.0000 23.914 .0836 .0418 .9198 34 3.0000 23.914 .1254 .0418 .8821 35 11.000 23.914 .4600 .0418 .6313 36 27.000 23.914 1.1290 .0418 .3233 37 35.000 23.914 1.4636 .0418 .2314 38 44.000 23.914 1.8399 .0418 .1588 39 49.000 115.73 .4234 .0086 .6548 40 12.000 46.612 .2574 .0215 .7730 41 27.000 46.612 .5793 .0215 .5603 42 117.00 46.612 2.5101 .0215 .0813 43 13.000 39.263 .3311 .0255 .7181 44 29.000 39.263 .7386 .0255 .4778 45 9.0000 35.231 .2555 .0284 .7746 46 49.000 35.231 1.3908 .0284 .2489 47 3.0000 63.171 .0475 .0158 .9536 48 28.000 63.171 .4432 .0158 .6420 49 104.00 63.171 1.6463 .0158 .1928 50 153.00 63.171 2.4220 .0158 .0887 51 61.000 72.733 .8387 .0137 .4323 52 25.000 41.537 .6019 .0241 .5478 53 10.000 21.830 .4581 .0458 .6325 54 3.0000 23.914 .1254 .0418 .8821 55 4.0000 23.914 .1673 .0418 .8460 56 22.000 23.914 .9199 .0418 .3985 57 32.000 23.914 1.3381 .0418 .2623 58 43.000 23.914 1.7981 .0418 .1656 59 100.00 23.914 4.1816 .0418 .0153 60 2.0000 46.612 .0429 .0215 .9580 61 21.000 46.612 .4505 .0215 .6373 62 38.000 46.612 .8152 .0215 .4425 --> SURVIVAL ; lhs = LogT ; Model = normal ; rhs = One, Prod $ Note: DFP and BFGS usually take more than 4 or 5 iterations to converge. If this problem was not structured for quick convergence, you might want to examine results closely. If convergence is too early, tighten convergence with, e.g., ;TLG=1.D-9. Normal exit from iterations. Exit status=0. +---------------------------------------------+ | Loglinear survival model: NORMAL | | Maximum Likelihood Estimates | | Model estimated: Nov 20, 2004 at 02:36:23AM.| | Dependent variable LOGT | | Weighting variable None | | Number of observations 62 | | Iterations completed 2 | | Log likelihood function -99.92876 | +---------------------------------------------+ +---------+--------------+----------------+--------+---------+----------+ |Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X| +---------+--------------+----------------+--------+---------+----------+ RHS of hazard model Constant 3.205656536 .17487529 18.331 .0000 PROD -9.180773817 3.0046309 -3.056 .0022 .11023065E-01 Ancillary parameters for survival Sigma 1.212659253 .14406937 8.417 .0000 +----------------------------------------------------------------+ | Parameters of underlying density at data means: | | Parameter Estimate Std. Error Confidence Interval | | ------------------------------------------------------------ | | Lambda .04485 .00768 .0298 to .0599 | | P .82463 .09797 .6326 to 1.0167 | | Median 22.29709 3.81654 14.8167 to 29.7775 | | Percentiles of survival distribution: | | Survival .25 .50 .75 .95 | | Time 50.52 22.30 9.84 3.03 | +----------------------------------------------------------------+ --> SURVIVAL ; lhs = LogT ; Model = logistic ; rhs = One, Prod $ Normal exit from iterations. Exit status=0. +---------------------------------------------+ | Loglinear survival model: LOGISTIC | | Maximum Likelihood Estimates | | Model estimated: Nov 20, 2004 at 02:36:24AM.| | Dependent variable LOGT | | Weighting variable None | | Number of observations 62 | | Iterations completed 6 | | Log likelihood function -101.3403 | +---------------------------------------------+ +---------+--------------+----------------+--------+---------+----------+ |Variable | Coefficient | Standard Error |b/St.Er.|P[|Z|>z] | Mean of X| +---------+--------------+----------------+--------+---------+----------+ RHS of hazard model Constant 3.292277409 .16511190 19.940 .0000 PROD -9.543597412 3.0786043 -3.100 .0019 .11023065E-01 Ancillary parameters for survival Sigma .7084697587 .97316748E-01 7.280 .0000 (Note: E+nn or E-nn means multiply by 10 to + or -nn power.) +----------------------------------------------------------------+ | Parameters of underlying density at data means: | | Parameter Estimate Std. Error Confidence Interval | | ------------------------------------------------------------ | | Lambda .04129 .00662 .0283 to .0543 | | P 1.41149 .19389 1.0315 to 1.7915 | | Median 24.21755 3.88060 16.6116 to 31.8235 | | Percentiles of survival distribution: | | Survival .25 .50 .75 .95 | | Time 52.74 24.22 11.12 3.01 | +----------------------------------------------------------------+ --> DSTAT;Rhs=PROD,T$ Descriptive Statistics All results based on nonmissing observations. =============================================================================== Variable Mean Std.Dev. Minimum Maximum Cases =============================================================================== ------------------------------------------------------------------------------- All observations in current sample ------------------------------------------------------------------------------- PROD .110230645E-01 .463625284E-01 -.104430000 .742700000E-01 62 T 42.6774194 45.8406986 1.00000000 216.000000 62