Abstract
This paper objects to the arising problems due to fitting graphical chain models to multidimensional data sets. This multivariate statistical tool is used to cope with complex research questions concerning not only direct, but also indirect associations between the variables of interest. Due to this high complexity sensible strategies for fitting such models are required. Here, a data--driven selection strategy is discussed. Its application is illustrated for an empirical data example in detail.
Dokumententyp: | Paper |
---|---|
Fakultät: | Mathematik, Informatik und Statistik > Statistik > Sonderforschungsbereich 386
Sonderforschungsbereiche > Sonderforschungsbereich 386 |
Themengebiete: | 500 Naturwissenschaften und Mathematik > 510 Mathematik |
URN: | urn:nbn:de:bvb:19-epub-1467-8 |
Sprache: | Englisch |
Dokumenten ID: | 1467 |
Datum der Veröffentlichung auf Open Access LMU: | 04. Apr. 2007 |
Letzte Änderungen: | 04. Nov. 2020, 12:45 |