Abstract
Following an introduction, this paper has three main sections discussing in turn theoretical advances associated with distribution modelling, empirical advances and outstanding problems. The first of these sections is concerned with the elucidation of structures of families of models, entropy- maximizing methods, balancing factors, the deterrence function and the need of further desegregation. The second section is concerned with measurement, calibration and testing methods, forecasting and providing evaluation inputs. Finally, a number of outstanding problems are discussed relating to many of the above headings.