Model used to represent choice between two mutually exclusive options; for example, a commuter may decide to drive to work or to use public transport. The multinomial logit model is mathematically simple (Andrew and Meen (2006) Papers Reg. Sci. 85, 3), and is widely used, but imposes the restriction that the distribution of the random error terms is independent and identical over the alternatives, causing the cross-elasticities between all pairs of alternatives to be identical, and this can produce biased estimates. The nested logit model allows the error terms of pairs or groups of alternatives to be correlated, but the remaining restrictions on the equality of cross-elasticities may be unrealistic (Hunt (2000) Papers Reg. Sci. 40, 1). Logit models which allow different cross-elasticity between pairs of alternatives include: the paired combinatorial logit (Koppelman and Wen (2000) Transp. Res. B4); the cross-nested logit (Vovsha (1997) Ann. Transpt. Res. Board, Washington, D.C.), and the product differentiation model (Filippini (1999) Int. J. Econ. Bus. 2). The major weakness of logit models is the implication that the choice between any two groups of alternatives depends solely on the characteristics of the alternatives being compared, and not upon the characteristics of any other alternatives in the choice set.