Quasi-variances and extensions

The notion of quasi-variances, as a device for both simplifying and enhancing the presentation of categorical-predictor e�ects in statistical models, was developed in Firth and de Menezes (Biometrika, 2004, 65{80). The approach generalizes the earlier ideas of Ridout (GLIM Proceedings, 1989) and of Easton, Peto and Babiker (Statistics in Medicine, 1991 | ` oating absolute risk', which has become rather controversial in epidemiology). In this talk I will outline and exemplify the method to show how it can be useful, and discuss its extension to some other contexts such as parameters that may be arbitrarily scaled and/or rotated.