The UCF Department of Statistics and Data Science in partnership with Addition Financial is pleased to announce our 2022-23 Analytics Competition! Participants will analyze real-world financial data to gain actionable insights into the problem of credit card delinquency. By creating a rigorous and reproducible statistical analysis and conveying results in a form understandable to non-experts, students will gain valuable experience for future employment. And to top it off, there is a $3,000 grand prize!



 Addition Financial’s Collections department reviews credit cards for payment delinquencies and defaults. Credit card delinquency occurs when a cardholder misses their required minimum monthly payment. Banks, credit unions, and credit card issuing companies monitor payment delinquencies to hedge against potential risk and to minimize losses when possible. Addition Financial’s Collections team monitors four credit card delinquency queues; first payment default (when a new cardholder or an existing member with a new card type misses their first expected payment), 60 day delinquency, 90 day delinquency, and 120 day delinquency. Cards delinquent 120 days or more are considered non-performing loans and are submitted for charge-off. Once charged-off they are considered a loss to the company. 

The Collections team works directly with the cardholder in an event where a cardholder appears in any of the aforementioned payment delinquency queues to help them fulfill their debt obligations. 

Presently, the delinquency queues are based on a number of days delinquent counter that resets back to zero once the minimum required payment is made. The purpose of this competition is to determine which data points have relevance and build a data model that can predict whether a cardholder would become 70-89 days delinquent on their payment the following month and by what amount.



Addition Financial Credit Union (AFCU) seeks to automate Collections’ monitoring process based on available historical payment data on a per member and card basis. The predictive analytics model should determine if members’ credit cards would become 70-89 days delinquent the following month and by what amount. This predictive tool would help Collections team focus their efforts on helping members and mitigating any potential loss to the credit union.



Contest participants will be provided anonymized historical credit card payment data. 

Contest participants will also be provided with a data dictionary, describing each of the data points. It will be imperative for participants to have a clear understanding of the data that they are working with—as much of the data is industry specific. Explaining their findings in a clear & concise manner will be just as important as having accurate results. 

The dataset can be downloaded here:

Data Download Link


Contest Structure 

The primary goal of the contest is to create a measurable way to predict whether a credit card would become 70-89 days delinquent on their payment the following month and by what amount. 

All entries will be judged based on the following criteria: 

  • Content in the Project Report
  • Modeling Approach
  • Results and Conclusions
  • Presentation of results 

Project codes must be written in R, Python, or Microsoft SQL. 

Code file types should be compatible with the software used: (example: 

R file should be in “*.r” (.rmd preferred), Python in “*.py” (.ipynb preferred), Microsoft SQL file should be in “*.sql”). 

**Teams will be required to formally present their findings to the AFCU competition judges at the AFCU Headquarters located at 1000 Primera Blvd. Lake Mary, FL 32746.** 

Judges will need to replicate the analysis using the team’s code/model on a validation dataset. Teams must provide formal instructions on how to run their model. Judging will be based on the application of the data, the method(s) used to reach conclusions, the number of significant relationships or correlations discovered, and the presentation of their findings. 

It is crucial that contest participants communicate their findings in layman’s terms. Treat this assignment as if you were presenting to a group of decision makers that may not have a strong background in rigorous statistical analysis. 


Competition Documents

See the documents below for the official competition description, rules, and ongoing Q&A regarding the dataset.



1st Place Team: $3,000
2nd Place Team: $1,500
3rd Place Team: $500


Important Dates

Contest start date: October 7, 2022 (data files will be available on this date)

Final submissions due: January 9, 2023 (early submissions are encouraged)

Group presentations will be conducted between February 6, 2023 and February 17, 2023

UCF analytics dept. notification of top three winners: March 6, 2023


Eligibility and Team Format

The contest is open to current full-time and part-time UCF students. Students will submit as a team with at least 1 member and up to 5 members. To register please use the link below:

Registration Form

Contestants are strongly advised to consult with a faculty member for guidance in this competition.


Kick-off Event

The competition kick-off was hosted in-person by the UCF Department of Statistics and Data Science on October 7th, 2022.

Kick Off Presentation

Passcode: wMg#L+c0


Questions & Contact

AFCU encourages students to contact us at any time with questions concerning the dataset. Please direct any questions you may have to

All questions and answers will be shared with other competitors via the FAQ page.



The submission link is here:

Submission Link