The Technion Prediction Tournament

Organized by: Ido Erev, Eyal Ert, and Alvin E. Roth

Supported by the Max Wertheimer Minerva Center for Cognitive Processing and Human Performance

Submission deadline: Sept 1^st, 2008

Early registration until Aug 15, 2008

Competition results & Winners added on Sep 2, 2008

(updated instructions April 16, 2008; Suggestions added on Aug 11, 2008)  

Motivation and the basic idea

Classical behavioral decision research focuses on counter-examples to rational decision theory (like the Allais’ paradox), and simple models of sufficient conditions to these counter-examples (like original prospect theory, Kahneman & Tversky, 1979).  We believe that this focus was highly effective in establishing the importance of behavioral decision research; yet, the future of our discipline depends on our ability to advance beyond counter-examples (see Budescu et al., 1998.

The current set of competitions tries to take one step beyond counter-examples by focusing on well defined spaces of choice problems.  The set includes three independent competitions that will focus on three distinct choice tasks:

One shot decisions under risk (like the situations studied by Kahneman and Tversky, 1979) - Condition Description

One shot decisions from experience  (like the situations studied by Hertwig et al., 2004) - Condition E-sampling

Repeated decisions from experience (like the situations studied by Barron and Erev, 2003) - Condition E-repeated

We ran one set of experimental studies (the estimation set) and will soon run a second set (the competition set).  Both sets focus on choice between two alternatives, a safe choice and a risky choice:

Safe: Medium with certainty;

Risk: Low with probability p; High otherwise

A draft that summarizes the results of the estimation set can be found Here: Comp.pdf

The basic task:

            The participants in each competition will be allowed to study the results of the estimation set.  Their goal will be to develop a model that will predict the results of the competition set.  The model should be implemented in a computer program that reads the parameter of the problems (Medium, Low, p and High) as an input and predicts the proportion of risky choices as an output. Thus, we use the generalization criterion methodology (see Busemeyer & Wang, 2000) and the implied competitions can be described as simplified variants of the competition organized by Arifovic, McKelvey, and Pevnitskaya (2006).

Additional information concerning the competitions can be found in the following pages:

1. The "registration" and submission page explains the actions that have to be taken in order to participate in the tournaments.   

2. The “problem selection algorithm” page presents the algorithm that was used to select the problems in the estimation set, and will be used to select (on April x) the problems in the competition set. 

3. The “Method” page presents the experimental method.

4. The “aggregated results” page presents the problems that were studied in the estimation set and the aggregated results.

5. The “raw data” pages present the raw data for each of the participants.

6. The "competition rules" page explains the time schedule, and the required features of the submissions.  

7. The "baseline models" pages present examples of possible submissions to the competition.

GOOD LUCK!