Upload
vudiep
View
215
Download
2
Embed Size (px)
Citation preview
Literaturverzeichnis
Anderson. N.H.: Comment on an analysis of variance model for the assessment of' configural cue uti 1 ization in cl inical judgement. Psychological Bulletin 1969. 72. 1.63-65.
Armstrong. J.S •• Denniston. W.B.(Jr.) & Gordon. M.M.: The use of the decomposition principle in making judgments. Org. Beh. Hum. Perf •• 1975. 14. 257-263.
Aufsattler. W •• Oswald. M •• Geisler. W •• Grasshoff. U. Eine Anlyse richterlicher Entscheidungen Ober die Strafrestaussetzung nach § 57 I StGB. Monatsschrift f. Kriminologie. 1982. 65. 305-317.
Badahur. R.R.: A representation of the joint distribution of responses to -n- dichotomous items. In: Solomon. H. (Ed.): Studies in item analysis and prediction. Stanford University Press 1961. 158-169.
Birch. M.W. : Maximum likelihood in three-way contingency tables. J. Roy. Stat. Soc. (Ser. B.) 1963. 25. 220-233.
Birch. M.W.: The detection of partial association:the 2x2 case. J. Roy. Statist. Soc (Ser.B) 1964. 26. 313-324.
Birnbaum. A. & Maxwell. A.E.: Classification procedures based on Bayes-formula. Applied Statistics. 1960. 2. 152-169.
Birnbaum. M.H. : The devil rides again: Correlation as an index of fit. Psychological Bulletin 1973, 12. 4, 239-242.
Birnbaum, M.H. : Reply to the devils advocates: don't confound model testing and measurement. Psychological Bulletin 1974. ai. 11. 854-859.
Bishop. Y.M.M •• Fienberg, S.E. & Holland. P.W.: Discrete multivariate analysis: theory and practice. MIT Press, Cambridge, 1975.
138
Bishop. Y.M.M. & Mosteller, F.: Smoothed contingency-table analysis. In: Bunker,J.P. (Ed.) : The national halothane study. (Kap. IV-3). U.S. Government Printing Office, Washington D.C. 1969.
Bock. R.D. : Multivariate statistical methods in behavioral research. McGraw-Hi 11, New Y.ork, 1975.
Borcherding, K. : Entscheidungstheorie und Entscheidungshi1feverfahren fOr komp1exe Entscheidungssituationen. In : Ir1e. M. : Methoden und Anwendungen in der Marktpsychologie. Hogrefe. G5ttingen, 1983
Brown, R. V. & Lindley, D. V.I Improving judgment by reconciling incoherence. Theory And Decision, 1982, 14, 113-132.
Camerer, C. General conditions for the success of bootstrapping models. Org. Beh. Hum. Perf., 1981, 27, 411-422.
Christl. H.L. & Stock. S.: Simu1ationsuntersuchungen Ober das Verha1-ten verschiedener automatischer Diagnoseverfahren. In: Lange, H. J. & Wagner. G. (Hrsg.) Computerunterstuetzte arzt1iche Diagnostik, S. 269-288 Schattauer, Stuttgart, New York. 1973
Cole, G. F. & Talarico. S. M. : Second thoughts on parole. American Bar Association Journal. 1977. 63, 973-976.
Cox. D. R. : The Analysis of binary Data. Methuen. London. 1970.
Cox. D. R. I The analysis of multivariate binary data. App1. Stat •• 1972,21. 113-120.
Croft, D. J.: Is computerized diagnosis possible? Computers And Biomedical Research, 1972, ~, 351-367.
Darroch, J. N. Interaction in multi-factor contingency tables. J. Roy. Stat. Soc. B. 1962, 24. 251-263.
Darroch, J. N. & Ratcliff, D.: Generalized iterative scaling for loglinear models. Ann. Math. Statist., 1972, ~. 5. 1470-1480.
Dawes, R. M. The robust beauty of improper linear models in decision making. . American Psychologist. 1979. ~, 7. 571-582.
139
Dawes. R.M. & Corrigan. B. : Linear models in decision making. Psychological Bulletin 1974. 81. 2. 95-106.
DeDombal. F. T. & Horrocks. J. C.: Use of receiver operating characteristic <ROC) curves to evaluate computer confidence threshold and clinical performance in the diagnosis of appendicitis. Meth. Inform. Med.~ 1978. 17. 157-161.
Deming. W.E. & Stephan. F.F. : On a least squares adjustment of a sampled frequency table when the expected marginal totals are known. Ann. Math. Statist. 1940. 11. 427-444.
Dickey. J. M.: Smoothed estimates for multinomial cell probabilities. Ann. Math. Statist •• 1967. a2.561-566.
Domas. P. A. & Peterson. C. R.: Probabilistic information processing systems: Evaluation with conditionally dependent data. Org Beh. Hum. Perf •• 1972. Z. 77-85.
Edwards. W. : Dynamic decision theory and probabilistic information processing. Human Factors. 1962. ~. 59-73.
Einhorn. H.J.: Use of nonlinear. noncompensatory models as a function of task and amount of information. Org. Beh. Hum. Perf •• 1971. 2. 1-27.
Einhorn, H. J. & Hogarth. R.M.: Unit weighting schemes for decision making. Org. Beh. Hum. Perf •• 1975. 13. 171-192.
Fienberg. S.E.: The analysis af cross-classified categorial data. MIT-Press Cambridge. 1977
Fleiss. J. L •• Spitzer. R. L •• Cohen. J. & Endicott. J.: Three computer diagnosis-methods compared. Arch. Gen. Psych •• 1972. 27. 643.
Fraser. P. & Franklin. D.A.: Mathematical models for the diagnosis of liver disease. <Problems arising in the ~se of conditional probability theory.) Quarterly Journal Of Medicine. 1974. 43. 73-78.
Frey. D. : Informationssuche und Informationsbewertung bei Entscheidungen. Bern. 1981
Fryback. D. G.: Bayes' theorem and conditional non-independence of data in a medical diagnosis task. Technical Report. 1973.
140
Geisser. S. : Discrimination. allocatory and separatory. linear aspects. In: Van Ryzin. J.:Claseification and Clustering. Academic Press N.Y •• 1977.
Gettys. C.F. & Willke. T.A. : The application of Bayes'Theorem when the true data state is uncertain. Organ. Beh. Hum. Perf •• 1969. ~. 125-141.
Gilbert. E.S. : On discrimination using qualitative variables. J. Amer. Statist. Assoc •• 1968. 63. 1399-1412.
Goldberg. L. R. : Simple models or simple processes? Some research on clinical judgements. American Psychologist. 1968. 23. 483-496.
Goldberg. L. R. : The search for configural relationships in personality assessments: The diagnosis of psychosis vs. neurosis from the MMPI. Multivariate Behavioral Research. 1969. ~. 523-536.
Goldberg. L. R. : Man versus model of man. A rationale plus some evidence for a method of improving on clinical inferences. Psych. Bulletin. 1970. 73. 422-432.
Good. I. J.: On the estimation of small frequencies in contingency tables. J. Roy. Statist. Soc •• (Ser. B). 1956. 18. 113-124.
Good. I. J. : The estimation of probabilities. MIT Press. Cambrigde. Mass •• 1965.
Good. I. J. : A Bayesian significance test for multinomial distributions. J. Roy. Statist. Soc. B. 1967. 29. 399-431.
Goodman. L. A. : On methods for comparing contingency tables. J. Roy. Statist. Soc. A. 1963. 126. 94-108.
Goodman. L. A. Simultaneous confidence limits for cross-product ratios in contingency tables. J. Roy. Statist. Soc. B. 1964. 26. 86-1D2.
Goodman. L. A.I Interactions in multidimensional contingency tables. Ann. Math. Statist •• 1964. 35. 632-646.
Goodman. L. A.: The multivariate analysis of qualitative data: interactions among multiple classifications. J. Amer. Stat. Assoc •• 1970. 65. 226-256.
141
Goodman. L. A.: Analyzing qualitative / categorial Data: loglinear models and latent structure analysis. Addison-Wesley. London. 1978
Goodman. L. A. & Kruskal. W. H. : Measures of association for cross classification. Springer. Heidelberg. New York. 1979.
Gottfredson. S.D. & Gottfredson. D.M. : Screening for risk rison of methods. Criminal Justice and Behavior 1980. Z. 315-330
a compa-
Greist. J.H •• Gustafson. D.H •• Stauss. F.F •• Rowse. G.L •• Laughgren. T.P. & Chiles. J.A.: Suicide risk prediction: a new approach. Life Threatening Behavior. 1974. ~. 212-223.
Grizzle. J.E •• Starmer. C.F. & Koch. G.G. data by linear models. Biometrics 1969. 25. 489-504.
Analysis of categorial
Habbema. J.D.F •• Hilden. J. & Bjerregaard. B.: The measurement of performance in probabilistic diagnosis. Meth. Inform. Med •• 1978. 17. 4. 217-226.
Habbema. J.D.F. Hermans. J. & Van Der Burgt. A.T.: Cases of doubt in allocation problems. Biometrica. 1974. 61. 313-324.
Haberman. S.J.: Loglinear fit for contingency tables (Algorithm As51) Applied Stat. 1972. 21. 218-225.
Haberman. S.J. Printing multidimensional tables (Algorithm As. 57) Applied Stat. 1973. 21. 118-126.
Habermann. S.J. : The analysis of frequency data. The University of Chicago Press 1974.
Habermann. S. J. : Analysis of qualitative data. Academic Press. New York. San Francisco. London. 1978.
HeiS. R. : Psychologische Diagnostik: EinfGhrung und Oberblick. In: HeiS. R •• Groffmann. K.J. & Michel. L. : Handbuch der Psychologie Band 6. Hogrefe. G5ttingen. 1964
Herman. L.M.: Predicting the effectiveness of Bayesian classification systems. Psychometrica. 1966. 31. 341.
142
Hershman. R.L.: Optimal inference and a redundancy measure for overlapping data sets. Org. Beh. Hum. Perf. 1973. 10, 225-242.
Hilden, J. Habbema, J.D.F. & Bjerregaard, B.: The measurement of performance in probabilistic diagnosis (III). Methods based on continuous functions of the diagnostic probabilities. Meth. Inform. Med., 1978, 17, 238-246.
Hills, M. Discrimination and allocation with discrete data. Applied Statistics, 1967, 16, 237.
Kahneman. D., Slovic,P. & Tversky,A. (Eds.) tainity: heuristicts and biases. Cambridge University Press 1982
Kaiser, G. : Kriminologie.
Judgement under Uncer-
C.F. Muller, Heidelberg, Karlsruhe 1980
Keeney, R.L. & Raiffa, H. : Decisions with multiple objectives: preferences and value tradeoffs. Wiley, 1976
Keeney, R.L. : Decision Analysis: An overview. Operations Research, 1982, 30, 803-838
Kendall, M. G. & Stuart, A. The Advanced Theory Of Statistics. Vol.3 Charles Griffin & Co., London, 1976 (3-rd. Ed.).
Killion, R. A. & Zahn, D. A. : A Bibliography of contingency table literature 1900 to 1974. Int. Stat. Rev., 1976, 44, 71-112.
Kritzer. H.H.: Introduction to multivariate contingency table analysis. American Journal Of Political Science, 1978, 22.
Kritzer, H. M. Nonmet II: A program for the analysis of contingency tables ••• by weighted least squares. Behavior Research, Methods And Instrumentation. 1976, ~, 320-321.
Ku, H.H. & Kullback, S.: Approximating discrete probability distributions. IEEE Transactions On Inf. Theory, 1969 IT 15. 444-447.
Ku, H.H •• Varner, R.N. & Kullback, S.: On the analysis of multidimensional contingency tables. J. Amer. Statist. Assoc., 1971, 66, 55-64.
Lachenbruch, P.A.: An almost unbiased method of obtaining confidence intervals for the probability of misc1assification in discriminant analysis. Biometrics, 1967, 23, 639-645.
Lachenbruch, P.A.: On expected probabilities of misc1assification in discriminant analysis, necessary sample size, and a relation with the multiple correlation coefficient. Biometrica, 1968, 24, 823-834.
Lachenbruch, P.A. & Mickey, M.R. : Estimation of error rates in discriminant analysis. Technometrics 1968, 10, 1-11.
Lancaster, H.O.: The Chi-squared distribution. Wiley, N.Y., 1969.
Lazarsfe1d, P.F.: Logical and mathematical foundation of latent structure analysis. In: Stouffer, S.A. et a1. (Eds.): Measurement and prediction. Princeton, University Press, 1950.
Lazarsfe1d, P.F.: The algebra of dichotomous systems. In: Solomon, H. (Ed.): Studies in item analysis and prediction. Stanford University Press, 1961.
Lichtenstein, S. Conditional non-independence of data in a practical Bayesian decision task. Org. Beh. Hum. Perf., 1972, e, 21-25.
Lichtenstein, S., Fischhoff, B. & Phillips, L. : Calibration of probabilities: the state of the art to 1980. In: Kahnemen, D., Slovic,P. & Tversky,A. (Eds.) : Judgement under Uncertainity: heuristicts and biases. Cambridge University Press 1982
Lindley, D. V., Tversky, A. & Brown, R. V. On the reconciliation of probability assessments. J. Roy. Stat. Soc. A., 1979, 1A2, Part 2, 146-180.
Lincoln, T.C. & Parker, R.D.: Medical diagnosis using Bayes Theorem. Health Service Research, 1967, 2, 34-45.
Ludwig, D. & Hei1bronn, D. : The design and testing of a new approach to computer-aided differential diagnosis. Meth. Inform. Med., 1983, 22, 156-166.
Macnaughton-Smith, P. : Entscheidungskriterien fOr vorzeitige Ent1assung aus der Haft (Parole>. Krimino1ogisches Journal, 1975, Z. 113-124.
144
Mannheim. H. & Wilkins. L. T. borstal training. London. 1955.
Prediction methods in relation to
Mathai. A. M. & Rathie. P. N. : Basic concepts in information theory and statistics. Wiley Eastern Limited. New Delhi. 1975
Medin. D. L •• Altom. M. W •• Edelson. S. M. & Freko. D. : Correlated symptoms an simulated medical classification. J. Exp. Psych.: Learning. Memory and Cognition. 1982. e. 37-50.
Meehl. P.E. Clinical vs. statistical prediction.
Meyer,
University Of Minnesota Press. Minneapolis, 1954.
F. : RUckfallprognose bei unbestimmt verurteilten Jugendlichen. Bonn, 1956.
Miller, G.A.: The magical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review. 1956, 63, 81-97.
Moore, H.D.: Evaluation of five discrimination procedures for binary variables. J. Amer. Statist. Assoc. 1973, 68, 399-404.
Mount. J.R. & Evans, J.W. : Computer aided diagnosis: a simulation study. Proceedings on the 5th IBM Medical Symposium 1963, 113.
Murphy, A. H. : A new vector partition of the probability score. J. Applied Meteorology, 1973, 12, 595-600.
Murphy, A. H. : Subjective quantification of uncertainity in real-time weather forecasts in the United States. In : Sz5ll5si-Nagy, A. & Wood, E.F. (Eds.) : Recent developments in real-time weather forecasting / control of weather resource systems. N.Y. Wiley, 1977.
Murphy, A. H. & Winkler, R. L. : Reliability of subjective forecasts of precipitation and temperature. Applied Statistics, 1977, 26, 41-47.
Navon, D.: The importance of being conservative: some reflections on human Bayesian behavior. Brit. J. Math. Stat. Psych., 1978, 31, 33-48.
145
Naylor, J.C. & Schenk, E.A.: The influence of cue redundancy upon the human inference process for tasks of varying degrees of predictability. argo Beh. Hum. Perf., 1968. ~, 47-61.
Neyman, J. : Contributions to the theory of the Chi-Square test. Berkley, Univ. of ,California Press. 1949.
Nichols, J. L. & Weinstein, E. B. : The effectiveness of education and treatment programs for drinking drivers: A decade of evaluation. In: Goldberg, L. (Ed.): Alcohol. drugs and traffic safety. Stockholm, 1981.
Nisbett, R. & Ross, L. : Human inference of human judgement.
strategies an shortcommings
Englewood Cliffs, N.Y. 1980.
Norusis, M. J. : Symptom non-independence in mathematical models for diagnosis. Phd. Dissertation. The University of Michigan. 1973.
Norusis, M.J. & Jacquez, J.A.: Symptom non-independence in mathematical models for diagnosis. Computers And Biomedical Research, 1975, ~, 156-172.
Raiffa, H. : Decision Analysis. Reading, Mass. Addison Wesley, 1968
Rossi, P. H., Berk, R. A. & Lenihan, K. J. Money, work and crime. New York: Academic Press, 1980.
Schaefer, R. E. & Borcherding, K. : A note on the consistency between two approaches to incorporate data from unreliable sources in Bayesian analysis. argo Beh. Hum. Perf., 1973. 2, 504-508.
Schaefer, R.E. : Probabilistische Informationsverarbeitung Methoden und ein Experiment. Bern, Huber 1976.
Theorie,
Schaefer, R. E. : Der Bayes-Ansatz in der diagnostischen Urteilsbildung. In: Jager, R. S., Mattenklott, A. & Schroder, R. D.: Diagnostische Urteilsbildung ind der Psychologie. Sonderdruck des Deutschen Instituts fOr Internationale Padagogische Forschung: Reihe: Studien zur Padagogischen Psychologie Bd. 20, 1984.
Schum, D.A. & Ducharme, W.M.: Comments on the relationship between the impact and the reliability of evidence. argo Beh. Hum. Perf. 1971, Q. 111-131.
146
Schum, D. A.: Inference on the basis of conditionally non independent data. J. Exp. Psych., 1966, 72, 401-409.
Scott, J. F., Stahura, J. & Vandiver, R.: Parole board decision making: An examination of the criteria utilized. In: Flynn, C. & Cornrad, J. (Eds.): The new and old criminology. New York, 1978, 285-298.
Sawyer, J. : Measurement and prediction, clinical and statistical. Psych. Bulletin, 1966, 66, 178-200.
Smith, J.E.K. : Analysis of qualitative data. Annual Review Of Psychology, 1976, 487-499.
Steiger, J.H. & Gettys, C.F.: Conditional dependence and decomposition strategies in diagnostic inference systems. Org. Beh. Hum. Perf., 1973, 10, 88-107.
Templeton, A.W., Jansen, C., Lehr, J.L. & Hufft, R.: Solitary pulmonary lesions. Computer-aided differential diagnosis and evaluation of mathematical methods. Radiology, 1967, 89, 605-613, Syracuse.
Tukey, J.W. : Exploratory data analysis. Addison Wesley, Reading, Mass., 1971.
Victor, N.: Probabilistische Zuordnungsverfahren. In: H. J. Lange & G. Wagner (Hrsg.): ComputerunterstGtzte arztliche Diagnostik. Schattauer, Stuttgart, New York, 1973.
Victor, N., Trampisch, H.J. & Zentgraf, R.: Diagnostic rules for qualitative variables with interactions. Meth. Inform. Med., 1974, 13, 184-186.
Victor, N.: Probleme der Auswahl geeigneter Zuordnungsregeln bei unvollstaendiger Information, insbesondere bei kategorialen Daten. Biometrics, 1976, 32, 571-585.
Wainer, H. : Estimating coefficients in linear models: It don't make no nevermind. Psych. Bulletin, 1976, 83, 213-217.
Wichmann, H.E., Koppen, L., Spechtmeyer, H. & Gross, R. : Zur Problematik des angemessenen Klassifikationsverfahrens. Meth. Inform. Med. 1978, 17, 47-54.
147
Yates. J. F. : External correspondence: Decomposition of the mean probability score. Org. Beh. Hum. Perf •• 1982. 30. 132-156.
Yntema. D.B. & Torgerson. W.S. : Man-computer cooperation in decisions requiring common sense. IRE Transactions on Human Factors in Electronics 1961. HFE-2. 20-26.
Zentgraf. R. & Victor. N.: Some problems arising in the statistical treatment of diagnosis. Meth. Inform. Med •• 1978. 17. 10-14.
Anhang A
Nachfolgend sind die Prozentsatze Uber- und unterkonfidenter Schatzungen bei der Annahme der bedingten Unabhangigkeit und beim second-order Modell fUr die einzelnen Analyse-Einheiten der drei Datensatze angegeben. Die Prozentsatze beziehen sich auf den jeweiligen Anteil der Falle an der Gesamtstichprobe.
Legende:
Spalte:
-1- Prozentsatz Uberkonfidenter Schatzungen bei der Annahme bedingt unabhangiger Indikatoren
-2- Prozentsatz unterkonfidenter Schatzungen bei der Annahme bedingt unabhangiger Indikatoren
-3- Prozentsatz Uberkonfidenter Schatzungen beim second-order Modell
-4- Prozentsatz unterkonfidenter Schatzungen beim second-order Modell
Die als 'durchschnittliche' Analyseeinheit ausgewahlte Variablenkombination ist mit *** MID *** gekennzeichnet, die Variablenkombination mit dem graBen Prozentsatz Uberkonfidenter Schatzungen ist mit *** MAX *** gekennzeichet.
Die als ELIMINIERT gekennzeichneten Analyseeinheiten wurden entfernt, weil die Beziehungsstuktur der jeweiligen Variablen durch die Annahme der stochastischen Unabhangigkeit nahezu vollstandig aufgeklart werden konnte.
149
Datensatz 1: kriminelle Vorbelastung.
Abhangige Variable mit 2 Stufen. 5 Indikatoren
(Prozentsatze Gber- und unterkonfidenter Schatzungen)
1 2 3 4
39.9 3.3 1.1 5.1 37.5 4.7 0.7 13.5 42.8 2.7 1.5 5.8 49.3 1.5 1.7 2.7
** MAX ** 54.5 0.2 1.0 3.3 49.3 1.0 1.6 2.9 24.3 0.4 0.2 2.8 25.1 0.3 0.5 2.8
-- ELIMINIERT ----22.2----0.1----1.5----2.8-------------27.9 0.1 1.4 1.4 24.0 1.8 1.8 13.6 29.1 0.8 1.8 13.3
** MID ** 38.9 1.4 1.3 12.6 41.5 0.6 2.0 3.1
ELIMINIERT ---- 5.2----0.0----1.1----0.4-------------46.7 4.3 2.4 4.1 52.8 2.2 4.9 5.9 47.5 3.0 1.9 6.0 45.6 0.9 1.5 2.3
-- ELIMINIERT ---- 5.9----0.0----1.1----1.0-------------10.4 1.8 2.7 3.3
MITTELWERTE : 38.2 1.7 1.7 5.81
150
Datensatz 2: Wahl-Daten
Abhangige Variable mit 3 Stufen. 5 Indikatoren
(Prozentsatze Ober- und unterkonfidenter Schatzungen)
1 2 3 4
26.1 6.7 4.8 9.1
** MAX ** 28.1 5.9 3.4 7.0 23.3 4.6 3.2 8.2 19.6 5.8 3.7 6.5 17.7 7.9 3.7 6.5 15.2 5.6 3.5 5.1 19.5 7.1 6.1 9.5
** MID ** 20.4 6.1 5.7 9.1 17.1 6.8 4.2 7.1 16.6 6.2 6.6 8.8 16.6 6.0 7.5 9.5 10.9 5.6 5.5 7.1 15.4 7.7 7.5 10.7 21.4 6.1 5.3 10.5
------------------------------------------------------------MITTELWERTE . 19.1 6.3 5.1 8.2 .
------------------------------------------------------------
151
Datensatz 3: Suizid-Daten
Abhangige Variable mit 2 Stufen. 5 Indikatoren
(Prozenteatze aber- und unterkonfidenter Schatzungen)
------------------------------------------------------------1 2 3 4
------------------------------------------------------------12.4 9.3 16.0 55.9 18.3 50.3 17.3 50.3 4.4 8.8 7.7 56.2
10.8 42.0 12.4 38.7 6.4 54.1 6.2 54.1
17.0 43.6 12.9 40.5 28.4 48.5 24.0 48.5 24.7 47.9 9.3 46.4 26.3 42.5 18.8 42.5
** MID ** 23.5 42.8 14.2 42.8 14.7 47.7 9.8 45.1 14.9 54.6 17.3 56.4 22.7 43.6 25.5 43.8 12.9 41.8 9.8 41.8 10.3 5.9 8.0 47.2 29.6 49.5 19.1 52.1
** MAX ** 55.9 0.0 10.1 45.4 27.1 47.2 27.6 45.6 25.8 45.1 24.5 45.1 30.4 49.5 29.4 49.5 25.0 4.4 25.0 44.8
------------------------------------------------------------MITTELWERTE : 21.0 37.1 16.4 47.3
Anhang B
Die im Ergebnisteil nicht abgebildeten Tabellen des 'Wahl'datensatzes sind hier wiedergegeben. Tabelle Bl veranschaulicht die Ergebnisse bei der als 'durchschnittlich' eingestufen Variablenkombination, Tabelle B2 die Ergebnisse der Variablenkombination mit dem maximalen Prozentsatz uberkonfidenter Schatzungen.
Tabelle B1<a)
Prozentsatz der Falle, fur den uber- bzw.
unterkonfidente Schatzungen errechnet werden
FEHLERTYP
(G = 0.75)
UEBERKON
FIDENZ
UNTERKON
FIOENZ
(Wahl daten)
MOOELL
ANNAHME
C.INO.
2-NO ORO. ,
% FAELLE
AKZEPT. UNKLAR
8.09
0.33
10.52
4.17
VERWERF.
12.28
5.40
___________ 1 _____________________________ _
C. IND.
2-NO ORO.
, , 1.72
3.53
14.57
13.68
4.39
5.57 , ,
-----______ 1 ______ ------------------------,
153
Tabelle BHb)
Die prozentuale Zuordnung der Falle zu den drei
Bereichen 'AKZEPT.' / 'UNKLAR.' / 'VERWERF.'
MODELL
I
% FAELLE
AKZEPT. UNKLAR VERWERF. _________________ 1 _____________________________ _
OBSERVED
C. IND.
2-ND ORO.
I I
I
6.04
11.22
4.24
76.42
64.65
74.81
17.54
24.13
20.95
-----------------,------------------------------
Tabelle B2(a)
Prozentsatz der FaIle. fur den uber- bzw. unter
konfidente Schatzungen errechnet werden
(Wahl daten)
FEHLER MODELL- % FAELLE
TYP ANNAHME AKZEPT. UNKLAR VERWERF. I I
----------- -----______ ' ______ ------------------------, I I I I
UEBERKON- C. IND. 13.77 8.24 14.32
FIDENZ 2-ND ORO. 0.02 1.49 3.35 I I I -----------,-----______ 1 ______ ------------------------I . I' ,
UNTERKON-
FIDENZ
I I I
C. IND. 1.22
2-ND ORO. 2.83
10.05 4.63
10.87 4.14
I I
I I I -----------,-----------,------------------------------,
154
Tablle B2(b)
Die prozentuale Zuordnung der F~lle zu den drei
Bereichen ' AKZEPT,' / 'UNKLAR,' / 'VERWERF,'
MODELL
G = 0.75
OBSERVED
C,IND,
2-ND ORO.
% FAELLE
AKZEPT. UNKLAR
6.91
15.19
2.28
74.80
60.21
76.52
VERWERF.
18.29
24.60
21.20
I ______________________________ 1