A Framework to Interpret Nonstandard Log-Linear Models
DOI:
https://doi.org/10.17713/ajs.v36i2.323Abstract
The formulation of log-linear models within the framework of Generalized Linear Models offers new possibilities in modeling categorical data. The resulting models are not restricted to the analysis of contingency tables in terms of ordinary hierarchical interactions. Such models are considered as the family of nonstandard log-linear models. The problem that can arise is an ambiguous interpretation of parameters. In the current paperthis problem is solved by looking at the effects coded in the design matrix and determining the numerical contribution of single effects. Based on these results, stepwise approaches are proposed in order to achieve parsimonious models. In addition, some testing strategies are presented to test such (eventually non-nested) models against each other. As a result, a whole interpretation framework is elaborated to examine nonstandard log-linear models in depth.
References
Alba, R. D. (1988). Interpreting the parameters in log-linear models. In J. S. Long (Ed.), Common Problems/Proper Solutions (p. 258-287). Newbury Park, CA: Sage.
Azen, R., and Budescu, D. V. (2003). The dominance analysis approach for comparing predictors in multiple regression. Psychological Methods, 8, 129-148.
Benedetti, J. K., and Brown, M. P. (1978). Strategies for the selection of log-linear models. Biometrics, 34, 680-686.
Bishop, Y. M. M., Fienberg, S. E., and Holland, P. W. (1975). Discrete Multivariate Analysis: Theory and Practice. Cambridge, MA: MIT Press.
Bock, R. D. (1975). Multivariate Statistical Methods in Behavioral Research. New York: McGraw-Hill.
Breen, R. (1984). Fitting nonhierarchical and association log-linear models using GLIM. Sociological Methods and Research, 13, 77-107.
Brown, M. P. (1976). Screening effects in multidimensional contingency tables. Applied Statistics, 25, 37-46.
Budescu, D. V. (1993). Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin, 114, 542-551.
Edwards, D., and Havranek, T. (1985). A fast procedure for model search in multidimensional contingency tables. Biometrika, 72, 339-351.
Elliott, G. C. (1988). Interpreting higher order interactions in log-linear analysis. Psychological Bulletin, 103, 121-130.
Evers, M., and Namboodiri, N. K. (1979). On the design matrix strategy in the analysis of categorical data. Sociological Methodology, 10, 86-111.
Eye, A. von, and Niedermeier, K. E. (1999). Statistical Analysis of Longitudinal Categorical Data in the Social and Behavioral Sciences. Mahwah, NJ: Erlbaum.
Eye, A. von, and Spiel, C. (1996). Standard and nonstandard log-linear symmetry model for measuring change in categorical variables. The American Statistician, 50, 300-305.
Fienberg, S. (1979). A note on fitting and interpreting parameters in models for categorical data. Sociological Methodology, 10, 112-118.
Gilbert, G. N. (1981). Modeling Society: An Introduction to Log-Linear Models for Social Researchers. Boston, MA: George Allen and Unwin.
Goodman, L. A. (1970). The multivariate analysis of qualitative data: Interactions among multiple classifications. Journal of the American Statistical Association, 59, 319-352.
Goodman, L. A. (1971). The analysis of multidimensional contingency tables: stepwise procedures and direct estimation methods for building models for multiple classifications. Technometrics, 13, 33-61.
Greene, W. H. (2003). Econometric Analysis (5 ed.). London: Prentice Hall.
Holt, D. (1979). Log-linear models for contingency table analysis: On the interpretation of parameters. Sociological Methods and Research, 7, 330-336.
Jennrich, R. I. (1969). Asymptotic properties of non-linear least squares estimators. The Annals of Mathematical Statistics, 40, 633-643.
Kruskal, W. (1987). Relative importance by averaging over orderings. The American Statistician, 41, 6-10.
Linhart, H. (1988). A test whether two AIC’s differ significantly. South African Statistical Journal, 22, 153-161.
Linhart, H., and Zucchini, W. (1986). Model Selection. New York: Wiley.
Magidson, J., Swan, J. H., and Berk, R. A. (1981). Estimating nonhierarchial and nested log-linear models. Sociological Methods and Research, 10, 3-49.
Mair, P. (2006). Interpreting Standard and Nonstandard Log-Linear Models. New York, Münster: Waxmann.
Mair, P., and Eye, A. von. (in preparation). Application scenarios for nonstandard loglinear models.
McCullagh, P., and Nelder, J. A. (1989). Generalized Linear Models (2nd ed.). London, New York: Chapman and Hall.
Nelder, J. A., and Wedderburn, R. W. M. (1972). Generalized linear models. Journal of the Royal Statistical Society, Series A, 135, 370-384.
Rindskopf, D. (1990). Nonstandard log-linear models. Psychological Bulletin, 108, 150-162.
Rindskopf, D. (1999). Some hazards of using nonstandard log-linear models, and how to avoid them. Psychological Methods, 4, 339-347.
Rudas, T. (1988). Odds Ratios in the Analysis of Contingency Tables. Thousand Oaks, CA: Sage.
Shimodaira, H. (1997). Assessing the error probability of the model selection test. Annals of the Institute of Statistical Mathematics, 49, 395-410.
Shtatland, E. S., Kleinman, K., and Cain, E. M. (2002). One more time about r2 measures of fit in logistic regression. In Proceedings of Northeast SAS User Groups (NESUG 15). Buffalo, NY.
Vermunt, J. K. (1997). LEM: A General Program for the Analysis of Categorical Data. Tilburg University.
Vuong, Q. H. (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57, 307-333.
Wickens, T. P. (1989). Multiway Contingency Tables Analysis for the Social Sciences. Hillsdale, NJ: Lawrence Erlbaum.
Wilson, T. P. (1979). On not interpreting coefficients: Comment on Holt. Sociological Methods and Research, 8, 233-240.
Downloads
Published
How to Cite
Issue
Section
License
The Austrian Journal of Statistics publish open access articles under the terms of the Creative Commons Attribution (CC BY) License.
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and transmit an article, adapt the article and make commercial use of the article. The CC BY license permits commercial and non-commercial re-use of an open access article, as long as the author is properly attributed.
Copyright on any research article published by the Austrian Journal of Statistics is retained by the author(s). Authors grant the Austrian Journal of Statistics a license to publish the article and identify itself as the original publisher. Authors also grant any third party the right to use the article freely as long as its original authors, citation details and publisher are identified.
Manuscripts should be unpublished and not be under consideration for publication elsewhere. By submitting an article, the author(s) certify that the article is their original work, that they have the right to submit the article for publication, and that they can grant the above license.
