The first Stat-a-thon will be held jointly by UConn Statistics Department and NESS NextGen during the 33rd New England Statistics Symposium on May 15--17, 2019. Stat-a-thon is an statistical data science invention marathon. The trademark is owned by University of Connecticut. Anyone who has an interest in data science attends a Stat-a-thon to approach a real world data science problem, some of which are local, in new and innovative ways. It emphasizes the statistical aspects (insight, interpretation, significance, etc.) of data science problems that are often overlooked in many hackathons. NextGen is a newly-formed committee within NESS supporting the next generation of statisticians and data scientists for them to contribute to the betterment of the New England Statistical Society.
We would like to congratulate the winners of the Travelers Stat-a-thon 2019:
Online registration opens, data sets released online with instructions.
Individuals looking to join an assigned team should register by this date, and we will provide your team information no later than April 1st.
Teams or individual participants should register by this deadline; online registration will be closed at the end of the day.
Deadline for teams to submit their work for the panelist to review.
Finalist teams are selected and notified.
Finalist teams present to the review panel in sections of NESS conference.
Awards to winning teams at the closing ceremony.
There are two themes for this Stat-a-thon. You may choose one that is interesting to you. Related data sets are provided for each theme. You are encouraged to use related data from other sources.
For this theme, there are no true answers, and a team needs to select its own goal to work on. A team does not have to answer all the questions below. You may choose some questions to focus on, and it is welcome to propose new questions to work on.
Housing has a significant impact on Connecticut’s economy and ability to attract and retain people. In addition, as a result of the great recession, many homeowners have not realized an appreciation in equity that is typically realized from owning a home. How can we better understand the impacts of the accessibility and affordability of housing? What is the impact on Connecticut’s economy? How can we help identify affordable and accessible places to live? What factors might affect housing prices?
Challenge: Using the primary dataset below, combine it with additional data sources to find interesting insights, trends, correlations, relationships, or patterns in housing in Connecticut.
Primary dataset: Real estate sales data provided by the State of Connecticut.
(Data sets are provided by Tyler Kleycamp, Chief Data Officer, Office of Policy and Management, State of Connecticut)
For this theme, there are true answers, and a team should focus on proposing the best predictive model. The performance of a team will be mainly based on the predictive performance of the propose method measured by AUC and the quality of the code.
Challenge: Using historical policy data, create a retention model to predict those policies that are most likely to cancel as well as understand what variables are most influential in causing a policy cancellation.
Training dataset: 4 years of property insurance policies from 2013 to 2017.
Test dataset: Test data for property insurance policies.
Data description: Variable descriptions.
(Data sets are synthetic, provided by Travelers)
All teams should register online. If you already have a team or want to participate as an individual, please register using the following link.
Each team may have up to four team members, and only one registration form should be submit by each team with all names of the team members.
If you do not have a team but want to be a part of one, please use the following form to register. The organizers will try to match you up with similar participants.
All teams should submit their work by the deadline (04/26/2019). Teams are encouraged to create a Git Repository (e.g., Bitbucket, GitHub) to host their source codes and data information. However, this is not a review factor in the competition.
Connecticut Housing Theme: Each team should submit a report along with other produce such as program codes used in the analysis, software products, links to other data sources if used, etc. The report should be in the format of presentation slides of up to ten pages. Finalist teams may extend their slides to more than ten pages for the presentation during the conference, but for the first round submission, the number of slides should not exceeds ten pages. A report should be submitted through TBA.
Customer Retention Theme: Teams working on this theme should submit their work through Kaggle in class using the link provided to you. No presentation slides are required in the first round submission. Finalist teams are expected to create slides based on their work and give presentations in a section of the NESS conference.
Ten teams (five from each theme) will be selected in the finalist, and they are invited to give team presentations to the review panels in the sessions of 05/16/2019. Travelers Insurance will be covering the conference registration fees of the NESS conference for all team members in the finalist. Each team will have twenty minutes to present their findings and products.
Students from universities and high schools can participate. We will not distinguish high school students, undergraduate students, and graduate students among participants.
No. Participation is free for the Stat-a-thon. We will select five finalist teams from each theme to come and present during the last day of the NESS conference. Travelers Insurance will be covering the conference registration fees for members of the finalist teams. However, team members may have to cover their own travel and/or lodging expenses.
Each team can have up to 4 participants.
Participants can form teams among peer students with common interests and/or complementary expertise. If you are not able to find a team yourself, you may either work individually, or request to be assigned with other participants that do not have a team. This is an opportunity for you to meet and work with new people. A participant can be a member of only one team.
You can start your work on the problem now.
Although we allow a team to work on both themes, we encourage you to focus on one theme in order to produce the best result. Note that each participant can only be in one team, so to work on both themes, the whole team must have an agreement. If you don't have a team, we will assign a team according to your preferred theme which means you may work on only one theme.
You can use any programming language or software packages.
Yes! There will be cash prizes for 1st, 2nd and 3rd place teams for both themes ranging from $500 to $100 dollars.
Direct links to the data sets for the Connecticut houseing theme are available on the Stat-a-thon website Theme 1. The customer retention theme will be utilizing Kaggle InClass for the stat-a-thon. You can download the data directly from the link provided on the Stat-a-thon website Theme 2.
Teams must be finalized no later than April 15th. If you are an individual looking to join an assigned team, you need to register before March 25 and we will provide that information to you no later than April 1st.
Yes, a professor or another professional can act as a team mentor. However, this person is not a member of the team, and cannot implement any work for the team.
Customer Retention Theme: Using the private Kaggle leaderboard, we will evaluate the teams that create the most accurate model score, compared to a gradient boosting machine model benchmark. The code of the top teams on the leaderboard will be reviewed, and based on the model score and code review, we will select 5 finalist teams. We are looking for each team to provide a business recommendation based on the results of your model.
Connecticut Housing Theme: Review panel will read team reports and codes to evaluate the completeness, validity, and innovation of your analysis. Since there are no true answer in this theme, we will focus on whether teams appropriately apply their statistical knowledge and skills to identify real problems, tell a complete story, provide interesting insights, develop novel solutions, and/or create intuitive and informative displays. We will select 5 finalist teams in this theme.
The 5 finalists from each challenge will be invited to present their work at the symposium and the winners will be selected among them.
HaiYing Wang (Chair), University of Connecticut
Tyler Kleykamp, Office of Policy and Managment, CT
Dooti Roy, Boehringer Ingelheim
Gregory Vaughan, Bentley University
Kathy Ziff, Travelers
For any further questions, please send them to