Analiza struktury zależności w tablicach kontyngencji za pomocą modeli logarytmiczno-liniowych

• Authors: Maria Sarama

Title variants

Analysis of Dependence Structure in Contingency Tables with Support of Logarithmic and Linear Models

• Languages of publication: PL

Abstracts

Omówiono wykorzystanie modeli logarytmiczno-liniowych do analizy tablic wielowymiarowych. Zastosowano je do analizy danych z publicznie dostępnej bazy German Credit Data. Zawiera ona informacje dotyczące 1000 wniosków kredytowych, obejmujące charakterystykę klientów i ocenę kredytu.

EN

This article concerns Log Linear models and its application to description and analysis of relationships in multidimensional contingency tables. The log-linear model and its parameters are defined. Methods and coefficients for selection and evaluation of models are presented. The theory is illustrated by an example based on database German Credit Data, data set taken from Machine Learning Database Repository, which describes good and bad bank customers.(original abstract)

Keywords

PL

Analiza wielowymiarowa; Analiza statystyczna; Metody statystyczne; Ocena zdolności kredytowej

EN

Multi-dimensional analysis; Statistical analysis; Statistical methods; Credit rating

• Publisher: Główny Urząd Statystyczny
• Journal: Wiadomości Statystyczne. The Polish Statistician
• Year: 2004
• Issue: nr 5
• Pages: 25-37

Contributors

author

Maria Sarama

References

• [1] Agresti A., Categorical Data Analysis, Wiley, New York 1990
• [2] Benedetti J.K., Brown M.B., Strategies for the Selection of Log Linear Models. Biometrics, 34 (1978), s. 680—686
• [3] Birch M.W., Maximum Likelihood in Three-way Contingency Tables, Journal of the Royal Statistical Society, ser. B, 25 (1963), s. 220—233
• [4] Bishop Y.M.M., Fienberg S.E., Holland P.W., Discrete Multivariate Analysis: Theory and Practice, MIT Press, Cambridge 1975
• [5] Brown M.B., Screening Effects in Multidimensional Contingencv Tables, Applied Statistics, 25 (1976).s, 34—46
• [6] Christensen R., Lug Linear Models and Logistic Regression, Springer, New York 1997
• [7] Darroch J.N., Interactions in Multifactor Contingency Tables, Journal of the Royal Statistical Society, ser. B, 24(1962), s. 162—166
• [8] Deming W.E.. Stephan F.F., On a Least Squares Adjustment of a Sampled Frequency Table When the Expected Marginal Totals Are Known, Annals of Mathematical Statistics, 11 (1944), s. 427—444
• [9] Goodman L.A., The Analysis of Multidimensional Contingency Tables: Stepwise Procedures and Direct Estimations Methods for Building Models for Multiple Classifications, Technometrics, 13 (1971), s. 33—61
• [10] Goodman L.A., Analyzing Qualitative/Categorical Data, Log Linear Models and Latent Structure Analysis, Abt Associates, Cambridge 1978
• [11] Haberman S.J., Log Linear Models for Frequency Data: Sufficient Statistics and Likelihood Equations. Annals of Statistics, 1(1973), s. 617—632
• [12] Imrey P.B., Koch G.G., Stokes M.E., Categorical Data Analysis: Some Reflections on the Log Linear Model and Logistic Regression. Part I: Historical and Methodological Overview, International Statistical Review, 49 (1981), s. 256—83
• [13] Imrey P.B., Koch G.G., Stokes M.E., Categorical Data Analysis: Some Reflections on the Log Linear Model and Logistic Regression. Part II: Data analysis. International Statistical Review, 50 (1982).
• s. 35—65