Comment on Prelec's (1998). "The Probability Weighting Function".

An alternative condition, called reduction invariance, is given to derive Prelec's (1998) form for the weighting function in separable utility theory. The condition, which is a variant on the reduction of compound gambles, is appreciably simpler and more empirically testable than Prelec's compound invariance and a simpler proof is provided. Both this and Prelec's condition are generalized leading to more general functions. Howener, of these various families, the only one that includes the rational power function as a special case is Prelec's formula.