INTRODUCTION The railways in Britain have a long history of using price discrimination backed up with product differentiation to significantly increase revenue over what might be obtained in an undifferentiated market. Whilst not as sophisticated as the yield management systems widely used in the airline industry, rail ticketing strategies are continually evolving, with new products emerging, unsuccessful products discontinued and gradual refinement of others. In recent years, there has been increased interest in modelling competition between different ticket types. The re-organisation of the railway industry in Great Britain has provided a greater commercial incentive to operators to price differentiate in order to maximise the revenue from their franchises. The policy of moderation of competition has allowed limited on-track competition, largely based around overlapping franchises but also with service extensions and new entrants, and this has stimulated product development and hence interest in ticket choice. Partly in response to the greater commercialisation of the railway industry, particularly where here is a degree of market power, the regulatory bodies have taken a greater interest in the range of tickets offered and their associated prices, travel restrictions and availability (SRA, 2003). This paper reports on research which was conducted as part of an update to the Passenger Demand Forecasting Handbook (PDFH), which contains a forecasting framework and recommended demand parameters that are widely used in the railway industry in Great Britain (ATOC, 2002), and as part of a project to provide the Strategic Rail Authority with evidence on cross elasticities between ticket types for use in its review of how it regulates rail fares.