It was found that the best-fitting combination (by some distance) was CRRA with
Quiggin; that a little over 80% of the subjects were clearly RDEU while a little under
20% were clearly SEU; and there was one rather confused subject who could not be
classified. The Harless and Camerer tremble occurred with probability of a little over
1.0%. Crucially the mixture specification fitted significantly better than either, assuming
that all subjects were SEU or that all were RDEU.
Harrison and Rutstrom (2009) also estimate a Mixture Model in a similar context,
using data from an experiment with 60 pairwise-choice questions. They too chose ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month, and much more.