Please consider downloading the latest version of Internet Explorer
to experience this site as intended.
Tools Search Main Menu

Better Living Through Statistics

April 20-21, 2018 at the University of Rochester, New York


Data Competition

The Statistical Data Analysis Competition provides a golden opportunity for students to explore new problem-solving techniques and fully showcase their statistical and mathematical skills on useful and practical real life problems. 

Prizes will be awarded to the top four teams, all of whom will be asked to present their results at the conference. Each team has only 10 minutes to express their results. This year's finalist teams are:

  1. Power Analysis Rangers
  2. Insight
  3. The Soothsayers
  4. StatsROCker

This year's challenge
A competition inspired by the Rochester City Police Department.

RPD Open Data Portal

Dates to Remember

  1. Registration Deadline: Mar 22
  2. Web access opens to download data: February 19
  3. Deadline to receive analysis and report: March 30

Register Your Team
Email with your team members’ names and email addresses. Faculty may advise a team, but the work would be done by students.

Submit Final Report
Details of submission in the PDF file here.

More Information

Team Constitution
Each team consists of a minimum of 3 and a maximum of 5 members. This competition is open to undergraduate and graduate students from all around the US.

Registration / Entry
Teams will have to register by the registration deadline (see below) in order to access the data and the description of the required tasks. All registrations will be made by email to Participation is limited to the first 50 teams that register so it is important to get your team(s) registered as early as possible.

Expectations and Tasks
Once a team is registered, it will have access to the data (see the timeline below), and the essence of the tasks include amongst other things:

  • Provide meaningful analyses and summaries of the data, through compelling graphical displays and tables.
  • Explore the data to seek and highlight significant structures and patterns.
  • Seek, find, and reveal relationships among the data types and describe them clearly
  • Compute wherever possible some predictive measures that could provide deeper insights into the internal structure of the data
  • Teams are free to use any computer software of their choice

Each team's entry will be reviewed by three different referees using a scholarly publication review style. Upon receipt, each entry (team's report and computer code, if any) will be assigned to the panel to three independent reviewers chosen by the Competition Committee. The merit of each entry will be measured according to the following five criteria

  1. [A] Novelty, Appeal and Compelling nature of the proposed solution (20 points)
  2. [B] Thoroughness/Completeness of the proposed solution (20 points)
  3. [C] Accuracy and Authenticity of the method(s) used (20 points)
  4. [D] Clarity, readability and overall professional quality of the final report (20 points)
  5. [E] Usefulness, usability, practicality and insights from the conclusions presented (20 points)

A leaderboard will be displayed on the conference website upon determination of the scores of each team. The top three teams will receive both prize money and certificates.  All the teams will get a certification of participation.

  1. Champion - Winner (Best team) – Largest Prize Money
  2. Runner up (second best team) – Second largest prize money
  3. 2nd Runner up (third best team) – Third prize money

All three top teams will be featured in a special AWARDS ceremony on the RIT Campus where they will present their final report orally with slides prior to the presentation of the prizes.

Format of Submission
The reviews will be kept anonymous. To help with that, teams are required to submit two versions of their report, one with their affiliations and the other without their affiliations. Since the quality of the report is one of the merit criteria, the Competition Committee will let each team decide the best way to prepare the report. However, the report must not exceed 10 (ten) pages. Each team is required to submit their computer code, if any, along with their report. At the discretion of the panel or the reviewer, the code may be run to verify the authenticity of the results claimed and presented. 

Format of Assessment
Reviewers’ reports must indicate clearly cases of technical/scientific incorrectness, along with the scores for each of the above 5 scoring categories, and a final recommendation. Once the first round of reviews is completed, the teams deemed worthy of further consideration will be identified and a second round of reviews may be performed to further refine the ranking of the teams.