home *** CD-ROM | disk | FTP | other *** search
File List | 1992-03-12 | 34.1 KB | 779 lines |
- SPSS/PC+ The Statistical Package for IBM PC 10/2/90
- INC '\spss\advtest.inc'
- title 'Test Job for SPSS/PC+ ADVANCED STATS'.
- SET LOG OFF VIEW 24.
- -------------------------------------------------------------------------------
- Page 2 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- DATA LIST /CASEID 1-2 SURVIVAL 4 TREATMNT 6 TIME 8-10(1)
- WEIGHT 12-15(3) APGAR 17-18 SEX 20 AGE 22-23
- PO2 25-27 PCO2 29-31 PH 33-35(2) RESP 37.
- VAR LABELS SURVIVAL 'Infant survival'
- TREATMNT 'Treatment administered'
- TIME 'Time to spontaneous respiration'
- WEIGHT 'Birthweight in kilograms'
- APGAR 'APGAR score'
- SEX 'Sex of respondent'
- AGE 'Gestation age'
- PO2 'O2 level'
- PCO2 'CO2 level'
- PH 'PH level'
- RESP 'Respiratory level'.
- RECODE SURVIVAL (0=1)(1=2).
- VALUE LABELS SURVIVAL 1 'Die' 2 'Survive'/
- TREATMNT 1 'THAM' 0 'Sodium bicarbonate'/
- SEX 1 'Male' 0 'Female'/
- RESP 1 'Yes' 0 'No' 9 'No answer'/.
- MISSING VALUES PCO2(-1)/RESP(9).
- BEGIN DATA
- END DATA.
- 50 cases are written to the compressed active file.
-
- This procedure was completed at 14:29:09
- -------------------------------------------------------------------------------
- Page 3 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- * next procedure requires that M module be installed.
-
- DSCRIMINANT GROUPS=SURVIVAL(1,2)
- /VARIABLES=TREATMNT TO AGE PH RESP
- /METHOD=DIRECT
-
- Since ANALYSIS= was omitted for the first analysis all variables
- on the VARIABLES= list will be entered at level 1.
- /STATISTICS=13 15
- /OPTIONS= 6 7.
-
-
- This Discriminant Analysis requires 2240 ( 2.2K) BYTES of workspace.
- -------------------------------------------------------------------------------
- Page 4 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- - - - - - - - - D I S C R I M I N A N T A N A L Y S I S - - - - - - - -
-
- On groups defined by SURVIVAL Infant survival
-
- 50 (unweighted) cases were processed.
- 2 of these were excluded from the analysis.
- 0 had missing or out-of-range group codes.
- 2 had at least one missing discriminating variable.
- 48 (unweighted) cases will be used in the analysis.
-
- Number of Cases by Group
-
- Number of Cases
- SURVIVAL Unweighted Weighted Label
- 1 26 26.0 Die
- 2 22 22.0 Survive
- Total 48 48.0
- -------------------------------------------------------------------------------
- Page 5 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- - - - - - - - - D I S C R I M I N A N T A N A L Y S I S - - - - - - - -
-
- On groups defined by SURVIVAL Infant survival
-
- Analysis number 1
- Direct method: All variables passing the tolerance test are entered.
- Minimum Tolerance Level.................. .00100
-
- Canonical Discriminant Functions
-
- Maximum number of functions.............. 1
- Minimum cumulative percent of variance... 100.00
- Maximum significance of Wilks' Lambda.... 1.0000
-
- Prior probability for each group is .50000
- -------------------------------------------------------------------------------
- Page 6 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- Canonical Discriminant Functions
-
- Pct of Cum Canonical After Wilks'
- Fcn Eigenvalue Variance Pct Corr Fcn Lambda Chisquare DF Sig
- : 0 .6150 20.419 8 .0089
- 1* .6261 100.00 100.00 .6205 :
-
- * marks the 1 canonical discriminant functions remaining in the analysis.
-
- Standardized Canonical Discriminant Function Coefficients
-
- FUNC 1
- TREATMNT .21531
- TIME .13170
- WEIGHT 1.16789
- APGAR .34638
- SEX .00337
- AGE -.64084
- PH .16862
- RESP -.58743
- -------------------------------------------------------------------------------
- Page 7 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- Structure Matrix:
-
- Pooled-within-groups correlations between discriminating variables
- and canonical discriminant functions
- (Variables ordered by size of correlation within function)
-
- FUNC 1
- WEIGHT .73338
- RESP -.51940
- AGE .49290
- TREATMNT .26572
- PH .25946
- APGAR .19210
- TIME -.10157
- SEX .03738
- -------------------------------------------------------------------------------
- Page 8 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- Canonical Discriminant Functions evaluated at Group Means (Group Centroids)
-
- Group FUNC 1
- 1 -.71252
- 2 .84206
-
- Symbols used in Plots
-
- Symbol Group Label
- ------ ----- --------------------
- 1 1 Die
- 2 2 Survive
- -------------------------------------------------------------------------------
- Page 9 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- All-groups stacked Histogram
-
- Canonical Discriminant Function 1
- 4 ┼ 2 1 2 ┼
- │ 2 1 2 │
- │ 2 1 2 │
- │ 2 1 2 │
- 3 ┼ 1 11222 2 2 ┼
- │ 1 11222 2 2 │
- │ 1 11222 2 2 │
- │ 1 11222 2 2 │
- 2 ┼ 11 211 11111222222 ┼
- │ 11 211 11111222222 │
- │ 11 211 11111222222 │
- │ 11 211 11111222222 │
- 1 ┼ 11 1111 111112211222 22 2 ┼
- │ 11 1111 111112211222 22 2 │
- │ 11 1111 111112211222 22 2 │
- │ 11 1111 111112211222 22 2 │
- X────┼────┼────┼────┼────┼────┼────┼────X
- Out -3.0 -2.0 -1.0 .0 1.0 2.0 3.0 Out
- Class 11111111111111111111122222222222222222222
- Centroids 1 2
- -------------------------------------------------------------------------------
- Page 10 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- Classification Results -
-
- No. of Predicted Group Membership
- Actual Group Cases 1 2
- -------------------- ------ -------- --------
- Group 1 26 22 4
- Die 84.6% 15.4%
-
- Group 2 22 4 18
- Survive 18.2% 81.8%
-
- Percent of "grouped" cases correctly classified: 83.33%
-
- Classification Processing Summary
- 50 Cases were processed.
- 0 Cases were excluded for missing or out-of-range group codes.
- 2 Cases had at least one missing discriminating variable.
- 48 Cases were used for printed output.
- -------------------------------------------------------------------------------
- Page 11 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- This procedure was completed at 14:29:23
-
- * next procedure requires that D module be installed.
-
- LOGISTIC REGRESSION SURVIVAL WITH TREATMNT AGE SEX WEIGHT APGAR.
-
-
- -------------------------------------------------------------------------------
- Page 12 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- Total number of cases: 50 (Unweighted)
- Number of selected cases: 50
- Number of unselected cases: 0
-
- Number of selected cases: 50
- Number rejected because of missing data: 0
- Number of cases included in the analysis: 50
-
-
-
- Dependent Variable Encoding:
-
- Original Internal
- Value Value
- 1 0
- 2 1
- -------------------------------------------------------------------------------
- Page 13 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- Dependent Variable.. SURVIVAL Infant survival
-
- Beginning Block Number 0. Initial Log Likelihood Function
-
- -2 Log Likelihood 68.994376
-
- * Constant is included in the model.
-
-
- Beginning Block Number 1. Method: Enter
-
- Variable(s) Entered on Step Number
- 1.. TREATMNT Treatment administered
- AGE Gestation age
- SEX Sex of respondent
- WEIGHT Birthweight in kilograms
- APGAR APGAR score
-
- Estimation terminated at iteration number 4 because
- Log Likelihood decreased by less than .01 percent.
-
- -------------------------------------------------------------------------------
- Page 14 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- Chi-Square df Significance
- -2 Log Likelihood 53.751 44 .1489
- Model Chi-Square 15.243 5 .0094
- Improvement 15.243 5 .0094
- Goodness of Fit 44.541 44 .4489
-
- Classification Table for SURVIVAL
- Predicted
- Die Survive Percent Correct
- D │ S
- Observed ┼────────┼────────┼
- Die D │ 22 │ 5 │ 81.48%
- ┼────────┼────────┼
- Survive S │ 9 │ 14 │ 60.87%
- ┼────────┼────────┼
- Overall 72.00%
-
- -------------------------------------------------------------------------------
- Page 15 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
- ---------------------- Variables in the Equation -----------------------
-
- Variable B S.E. Wald df Sig R Exp(B)
-
- TREATMNT .8414 .6804 1.5295 1 .2162 .0000 2.3197
- AGE -.2907 .2419 1.4447 1 .2294 .0000 .7477
- SEX -.4179 .7677 .2963 1 .5862 .0000 .6584
- WEIGHT 3.1243 1.3639 5.2472 1 .0220 .2169 22.7438
- APGAR .0704 .1352 .2708 1 .6028 .0000 1.0729
- Constant 2.7913 5.8809 .2253 1 .6350
- -------------------------------------------------------------------------------
- Page 16 Test Job for SPSS/PC+ ADVANCED STATS 10/2/90
-
- This procedure was completed at 14:29:38
-
- * next procedure requires that E module be installed.
- TITLE 'PROBIT Analysis'.
- DATA LIST / Root 1 X 3-7 N 9-11 R 13-15.
- BEGIN DATA
- END DATA.
- 10 cases are written to the compressed active file.
-
- This procedure was completed at 14:29:42
-
- PROBIT R of N by ROOT(1,2) WITH X
- /MODEL = BOTH
- /NATRES
- /PRINT ALL.
- PROBIT requires 776 BYTES of workspace for execution.
- -------------------------------------------------------------------------------
- Page 17 PROBIT Analysis 10/2/90
-
- * * * * * * * * * * * P R O B I T A N A L Y S I S * * * * * * * * * * *
-
- DATA Information
-
- 9 unweighted cases accepted.
- 0 cases rejected because of out-of-range group values.
- 0 cases rejected because of missing data.
- 0 Cases rejected because LOG-transform can't be done.
-
-
- Group Information
-
- ROOT Level N of Cases Label
- 1 4 1
- 2 5 2
-
-
- MODEL Information
-
- BOTH Probit and Logit models are requested.
-
-
- Natural Response rate to be estimated
-
- The number of subjects in the CONTROL group 129.0
- The number of responses in the CONTROL group 21.0
- -------------------------------------------------------------------------------
- Page 18 PROBIT Analysis 10/2/90
-
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
- -------------------------------------------------------------------------------
- Page 19 PROBIT Analysis 10/2/90
-
- LOG(X) VS. LOGIT
-
- -+----+----+----+----+----+----+----+----+----+----+----+----+--
- : :
- : 1 :
- L : 2 :
- O : :
- G 6 + 2 1 +
- I : 2 :
- T : :
- : 1 :
- : :
- 4 + 2 +
- : :
- -+----+----+----+----+----+----+----+----+----+----+----+----+--
- .45 .75 1.05 1.35 1.65 1.95 2.25
- .6 .9 1.2 1.5 1.8 2.1
-
- LOG(X)
- -------------------------------------------------------------------------------
- Page 20 PROBIT Analysis 10/2/90
-
- ML converged at iteration 6. The converge criterion = .00095
-
- Parameter Estimates (LOGIT model: (LOG(p/(1-p))/2 + 5) = Intercept + BX ):
- Note 5 added to intercept and logit divided by 2.
-
- Regression Coeff. Standard Error Coeff./S.E.
-
- X 2.45569 .19850 12.37135
-
-
- Intercept Standard Error Intercept/S.E. ROOT
-
- 2.02382 .30064 6.73182 1
- 2.59473 .28483 9.10988 2
-
- Estimate of Natural Response Rate = .172976 with S.E. = .03147
-
- Pearson Goodness-of-Fit Chi Square = 8.966 DF = 6 P = .176
- PARALLELISM TEST CHI SQUARE = .048 DF = 1 P = .826
-
- Since Goodness-of-Fit Chi square is NOT significant, no heterogeneity
- factor is used in the calculation of confidence limits.
- -------------------------------------------------------------------------------
- Page 21 PROBIT Analysis 10/2/90
- * * * * * * * * * * * P R O B I T A N A L Y S I S * * * * * * * * * * *
- Covariance(below) and Correlation(above) Matrices of Parameter Estimates
-
- X NAT RESP
-
- X .03940 .42344
- NAT RESP .00264 .00099
- -------------------------------------------------------------------------------
- Page 22 PROBIT Analysis 10/2/90
-
- Observed and Expected Frequencies
-
-
- Number of Observed Expected
- ROOT X Subjects Responses Responses Residual Prob
-
- 1 2.17 142.0 142.0 140.728 1.272 .99105
- 1 2.00 127.0 126.0 124.406 1.594 .97958
- 1 1.68 128.0 115.0 116.388 -1.388 .90928
- 1 1.08 126.0 58.0 43.158 14.842 .34252
-
- 2 1.79 125.0 125.0 122.735 2.265 .98188
- 2 1.66 117.0 115.0 113.057 1.943 .96630
- 2 1.49 127.0 114.0 117.491 -3.491 .92512
- 2 1.17 51.0 40.0 36.644 3.356 .71850
- 2 .58 132.0 37.0 16.255 20.745 .12314
-
- -------------------------------------------------------------------------------
- Page 23 PROBIT Analysis 10/2/90
-
- Confidence Limits for Effective X
-
-
- ROOT 1 1
-
- 95% Confidence Limits
- Prob X Lower Upper
-
- .01 1.88949 1.09495 2.83761
- .02 2.62750 1.61246 3.78991
- .03 3.19291 2.02582 4.49950
- .04 3.67174 2.38515 5.09032
- .05 4.09673 2.71020 5.60842
- .06 4.48451 3.01123 6.07679
- .07 4.84483 3.29438 6.50879
- .08 5.18404 3.56370 6.91298
- .09 5.50649 3.82200 7.29523
- .10 5.81537 4.07136 7.65976
- .15 7.22389 5.22909 9.30609
- .20 8.50530 6.30599 10.78770
- .25 9.73339 7.35390 12.19915
- .30 10.95051 8.40392 13.59378
- .35 12.18736 9.47954 15.00993
- -------------------------------------------------------------------------------
- Page 24 PROBIT Analysis 10/2/90
- .40 13.47086 10.60208 16.48107
- .45 14.82836 11.79372 18.04134
- .50 16.29114 13.08011 19.73008
- .55 17.89822 14.49337 21.59690
- .60 19.70188 16.07639 23.70910
- .65 21.77676 17.88994 26.16406
- .70 24.23644 20.02552 29.11216
- .75 27.26708 22.63118 32.80390
- .80 31.20422 25.96981 37.69896
- .85 36.73936 30.57370 44.76500
- .90 45.63792 37.77256 56.53912
- .91 48.19789 39.80239 60.01242
- .92 51.19587 42.15862 64.12498
- .93 54.78029 44.94814 69.10254
- .94 59.18183 48.33569 75.29989
- .95 64.78363 52.59221 83.31382
- .96 72.28220 58.20465 94.24584
- .97 83.12198 66.16875 110.42335
- .98 101.00910 78.99264 137.96693
- .99 140.46189 106.22774 201.78711
- -------------------------------------------------------------------------------
- Page 25 PROBIT Analysis 10/2/90
-
- Confidence Limits for Effective X
-
-
- ROOT 2 2
-
- 95% Confidence Limits
- Prob X Lower Upper
-
- .01 1.10627 .60305 1.73822
- .02 1.53837 .88826 2.32107
- .03 1.86941 1.11620 2.75507
- .04 2.14976 1.31446 3.11617
- .05 2.39859 1.49390 3.43263
- .06 2.62562 1.66017 3.71854
- .07 2.83659 1.81665 3.98208
- .08 3.03519 1.96556 4.22851
- .09 3.22398 2.10846 4.46142
- .10 3.40483 2.24647 4.68340
- .15 4.22950 2.88824 5.68417
- .20 4.97975 3.48680 6.58206
- .25 5.69878 4.07090 7.43472
- .30 6.41139 4.65794 8.27438
- .35 7.13555 5.26127 9.12395
- -------------------------------------------------------------------------------
- Page 26 PROBIT Analysis 10/2/90
- .40 7.88702 5.89319 10.00308
- .45 8.68182 6.56666 10.93157
- .50 9.53826 7.29687 11.93193
- .55 10.47919 8.10303 13.03228
- .60 11.53521 9.01097 14.27051
- .65 12.75003 10.05751 15.70113
- .70 14.19014 11.29835 17.40806
- .75 15.96454 12.82395 19.53073
- .80 18.26969 14.79521 22.32464
- .85 21.51044 17.53817 26.32751
- .90 26.72044 21.86628 32.95019
- .91 28.21927 23.09232 34.89717
- .92 29.97455 24.51759 37.20002
- .93 32.07319 26.20723 39.98453
- .94 34.65024 28.26152 43.44843
- .95 37.93002 30.84530 47.92442
- .96 42.32034 34.25466 54.02659
- .97 48.66690 39.09473 63.05261
- .98 59.13959 46.88897 78.41486
- .99 82.23871 63.43687 113.99793
- -------------------------------------------------------------------------------
- Page 27 PROBIT Analysis 10/2/90
-
- Estimates of Relative Median Potency
-
-
- 95% Confidence Limits
- ROOT Estimate Lower Upper
-
- 1 VS. 2 1.7080 1.31475 2.30747
- -------------------------------------------------------------------------------
- Page 28 PROBIT Analysis 10/2/90
-
- LOG(X) VS. PROBIT
-
- -+----+----+----+----+----+----+----+----+----+----+----+----+--
- : :
- : 2 1 :
- P : :
- R : :
- O 6 + 2 1 +
- B : 2 :
- I : :
- T : 1 :
- : :
- 4 + 2 +
- : :
- -+----+----+----+----+----+----+----+----+----+----+----+----+--
- .45 .75 1.05 1.35 1.65 1.95 2.25
- .6 .9 1.2 1.5 1.8 2.1
-
- LOG(X)
- -------------------------------------------------------------------------------
- Page 29 PROBIT Analysis 10/2/90
-
- ML converged at iteration 6. The converge criterion = .00063
-
- Parameter Estimates (PROBIT model: (PROBIT(p) + 5) = Intercept + BX ):
- Note 5 added to intercept.
-
- Regression Coeff. Standard Error Coeff./S.E.
-
- X 2.78382 .20654 13.47809
-
-
- Intercept Standard Error Intercept/S.E. ROOT
-
- 1.59842 .33212 4.81282 1
- 2.26158 .30366 7.44765 2
-
- Estimate of Natural Response Rate = .168731 with S.E. = .03177
-
- Pearson Goodness-of-Fit Chi Square = 5.933 DF = 6 P = .431
- PARALLELISM TEST CHI SQUARE = .050 DF = 1 P = .823
-
- Since Goodness-of-Fit Chi square is NOT significant, no heterogeneity
- factor is used in the calculation of confidence limits.
- -------------------------------------------------------------------------------
- Page 30 PROBIT Analysis 10/2/90
- * * * * * * * * * * * P R O B I T A N A L Y S I S * * * * * * * * * * *
- Covariance(below) and Correlation(above) Matrices of Parameter Estimates
-
- X NAT RESP
-
- X .04266 .46719
- NAT RESP .00307 .00101
- -------------------------------------------------------------------------------
- Page 31 PROBIT Analysis 10/2/90
-
- Observed and Expected Frequencies
-
-
- Number of Observed Expected
- ROOT X Subjects Responses Responses Residual Prob
-
- 1 2.17 142.0 142.0 141.411 .589 .99586
- 1 2.00 127.0 126.0 125.075 .925 .98485
- 1 1.68 128.0 115.0 115.136 -.136 .89950
- 1 1.08 126.0 58.0 43.540 14.460 .34556
-
- 2 1.79 125.0 125.0 123.477 1.523 .98782
- 2 1.66 117.0 115.0 113.566 1.434 .97065
- 2 1.49 127.0 114.0 116.994 -2.994 .92121
- 2 1.17 51.0 40.0 35.611 4.389 .69825
- 2 .58 132.0 37.0 17.215 19.785 .13042
-
- -------------------------------------------------------------------------------
- Page 32 PROBIT Analysis 10/2/90
-
- Confidence Limits for Effective X
-
-
- ROOT 1 1
-
- 95% Confidence Limits
- Prob X Lower Upper
-
- .01 2.43358 1.49073 3.54170
- .02 3.04907 1.93470 4.32588
- .03 3.51801 2.28205 4.91252
- .04 3.91772 2.58346 5.40658
- .05 4.27613 2.85742 5.84556
- .06 4.60690 3.11310 6.24774
- .07 4.91792 3.35578 6.62356
- .08 5.21417 3.58886 6.97966
- .09 5.49906 3.81465 7.32052
- .10 5.77505 4.03482 7.64936
- .15 7.07305 5.08714 9.18084
- .20 8.30973 6.11123 10.62212
- .25 9.54166 7.14783 12.04574
- .30 10.80286 8.22256 13.49459
- .35 12.11989 9.35637 15.00163
- -------------------------------------------------------------------------------
- Page 33 PROBIT Analysis 10/2/90
- .40 13.51780 10.56970 16.59764
- .45 15.02357 11.88506 18.31571
- .50 16.66914 13.32937 20.19508
- .55 18.49495 14.93677 22.28584
- .60 20.55513 16.75274 24.65563
- .65 22.92596 18.84086 27.40086
- .70 25.72100 21.29482 30.66690
- .75 29.12076 24.26177 34.68812
- .80 33.43794 27.99319 39.87663
- .85 39.28436 32.97398 47.05260
- .90 48.11392 40.33718 58.20447
- .91 50.52866 42.31938 61.31623
- .92 53.28948 44.57022 64.90463
- .93 56.49958 47.16756 69.11657
- .94 60.31388 50.22764 74.17419
- .95 64.97940 53.93463 80.43461
- .96 70.92392 58.60509 88.52216
- .97 78.98231 64.85142 99.66931
- .98 91.12942 74.10611 116.83393
- .99 114.17771 91.23085 150.43809
- -------------------------------------------------------------------------------
- Page 34 PROBIT Analysis 10/2/90
-
- Confidence Limits for Effective X
-
-
- ROOT 2 2
-
- 95% Confidence Limits
- Prob X Lower Upper
-
- .01 1.40612 .83171 2.10009
- .02 1.76176 1.07972 2.56433
- .03 2.03271 1.27386 2.91142
- .04 2.26367 1.44239 3.20359
- .05 2.47075 1.59564 3.46308
- .06 2.66188 1.73871 3.70072
- .07 2.84158 1.87455 3.92270
- .08 3.01275 2.00506 4.13295
- .09 3.17737 2.13153 4.33413
- .10 3.33683 2.25489 4.52814
- .15 4.08682 2.84506 5.43077
- .20 4.80137 3.42031 6.27872
- .25 5.51318 4.00355 7.11474
- .30 6.24190 4.60933 7.96390
- .35 7.00289 5.24966 8.84530
- -------------------------------------------------------------------------------
- Page 35 PROBIT Analysis 10/2/90
- .40 7.81060 5.93636 9.77656
- .45 8.68064 6.68260 10.77649
- .50 9.63145 7.50419 11.86722
- .55 10.68641 8.42131 13.07685
- .60 11.87678 9.46096 14.44321
- .65 13.24665 10.66101 16.02005
- .70 14.86163 12.07738 17.88828
- .75 16.82601 13.79805 20.17822
- .80 19.32048 15.97345 23.11903
- .85 22.69855 18.89332 27.16715
- .90 27.80029 23.23350 33.43063
- .91 29.19553 24.40516 35.17470
- .92 30.79073 25.73678 37.18464
- .93 32.64553 27.27463 39.54249
- .94 34.84944 29.08777 42.37235
- .95 37.54519 31.28554 45.87376
- .96 40.97994 34.05583 50.39566
- .97 45.63608 37.76201 56.62695
- .98 52.65470 43.25369 66.22123
- .99 65.97204 53.41329 85.00561
- -------------------------------------------------------------------------------
- Page 36 PROBIT Analysis 10/2/90
-
- Estimates of Relative Median Potency
-
-
- 95% Confidence Limits
- ROOT Estimate Lower Upper
-
- 1 VS. 2 1.7307 1.34843 2.30803
- -------------------------------------------------------------------------------
- Page 37 PROBIT Analysis 10/2/90
-
- This procedure was completed at 14:30:04
-
-
-
- Title Non-linear model for windspeed measurement.
-
- DATA LIST / X 1-3 Y 5-9.
- VAR LABELS X 'anomometer height' Y 'windspeed'.
- BEGIN DATA
- END DATA.
- 5 cases are written to the compressed active file.
-
- This procedure was completed at 14:30:07
-
- * next procedure requires that E module be installed.
-
- MODEL PROGRAM
- A=100 B=7.06 C=-223.26.
- -------------------------------------------------------------------------------
- Page 38 NON-LINEAR MODEL FOR WINDSPEED MEASUREMENT 10/2/90
-
- This procedure was completed at 14:30:11
- COMPUTE PRED=A*LN(B*X+C).
- NLR Y WITH X/PRED=PRED/SAVE PRED.
-
- All the derivatives will be calculated numerically.
- -------------------------------------------------------------------------------
- Page 39 NON-LINEAR MODEL FOR WINDSPEED MEASUREMENT 10/2/90
-
- There are 5 cases. There is enough memory for them all.
-
- Iteration Residual SS A B C
-
- 1 6822.703884 100.000000 7.06000000 -223.26000
- 1.1 3899.279813 107.235597 2.71055158 -18.043743
- 2 3899.279813 107.235597 2.71055158 -18.043743
- 2.1 107.8417730 115.003329 2.27328009 -24.273351
- 3 107.8417730 115.003329 2.27328009 -24.273351
- 3.1 7.040578209 115.150921 2.30999733 -22.101959
- 4 7.040578209 115.150921 2.30999733 -22.101959
- 4.1 7.013266157 115.147444 2.31056594 -22.026921
- 5 7.013266157 115.147444 2.31056594 -22.026921
- 5.1 7.013265848 115.146862 2.31064751 -22.028813
- 6 7.013265848 115.146862 2.31064751 -22.028813
- 6.1 7.013265848 115.146870 2.31064641 -22.028787
-
- Run stopped after 12 model evaluations and 6 derivative evaluations.
- Iterations have been stopped because the relative reduction between successive
- residual sums of squares is at most SSCON = 1.000E-08
- -------------------------------------------------------------------------------
- Page 40 NON-LINEAR MODEL FOR WINDSPEED MEASUREMENT 10/2/90
-
-
- Nonlinear Regression Summary Statistics Dependent Variable Y
-
- Source DF Sum of Squares Mean Square
-
- Regression 3 2314527.69673 771509.23224
- Residual 2 7.01327 3.50663
- Uncorrected Total 5 2314534.71000
-
- (Corrected Total) 4 75525.34800
-
- R squared = 1 - Residual SS / Corrected SS = .99991
-
- Asymptotic 95 %
- Asymptotic Confidence Interval
- Parameter Estimate Std. Error Lower Upper
-
- A 115.14686999 2.040556318 106.36706478 123.92667520
- B 2.310646412 .280311752 1.104562288 3.516730537
- C -22.02878660 6.409417923 -49.60628612 5.548712919
- -------------------------------------------------------------------------------
- Page 41 NON-LINEAR MODEL FOR WINDSPEED MEASUREMENT 10/2/90
-
- Asymptotic Correlation Matrix of the Parameter Estimates
-
- A B C
-
- A 1.0000 -.9963 .9666
- B -.9963 1.0000 -.9802
- C .9666 -.9802 1.0000
- -------------------------------------------------------------------------------
- Page 42 NON-LINEAR MODEL FOR WINDSPEED MEASUREMENT 10/2/90
-
- This procedure was completed at 14:30:20
- STOP.
-
- End of Include file.