August 2, 2018
July 29, 2018
The competition data (namely, what had to be predicted) is uploaded to the website.
Participants may check their MSE
Track I (Aggregate behavior):
Aggregate choice rates
Aggregate choice rates: Problems listed with complete distributions
Raw data
Track II (Individual behavior):
Individual choice rates by game and block
Raw data
July 25, 2018
Teams who submitted to Track I (Aggregate behavior):
Please download the competition problem set below (according to the programming language you used), read it into your code, and generate a csv file with your predictions for the 60 competition problems (one row per problem with predictions for the 5 blocks). You should then submit your csv of predictions through here (by 7/28)
For those who followed the structure of the code as requested (with a wrapper “main” file similar to that provided in the baseline codes), here is a very simple recipe for generating your predictions.
Competition set problems:
For R or Python
For SAS
For Matlab
July 22, 2018:
Code (Matlab) for the problem selection algorithm (i.e. for the pool of problems generating problems in Track I) has been added to website here
July, 18, 2018:
The rules of the competition state that each person may be the co-author in up to 2 submissions (per track), and be the first author in 1 of these. These rules led to some confusions in registrations. Therefore, we announce that each person currently registered to make only one submission (for a particular track) is now allowed to make 2 submissions to this track (even without additional registration). However, note that both rules stand (no more than 2 submissions per track, and only one first-author submission per track).