Structural breaks are a critical concept in time series analysis and econometrics. They refer to sudden changes in the underlying structure of a time series, which can significantly impact forecasting and modeling. Understanding and identifying structural breaks is essential for accurate predictions and informed decision-making in various fields, including finance, economics, and social sciences.
In Industry: Pattern-breaking sensor data can signal machine wear, the onset of failure, or process drift.
In Finance: A regime shift can invalidate entire risk models. Early detection enables dynamic portfolio rebalancing, volatility forecasting, and adaptive trading.
In Climate: Abrupt changes in temperature or pressure trends signal broader anomalies and potentially early warnings of irreversible change.
In Healthcare: A subtle shift in vital signs can mean the difference between stability and crisis.
In this competition, you are given univariate time series data which, at an explicitly given point, may have experienced a structural break, meaning that the process governing the data generation may have changed from that point on.
Participants have to develop methods that determine whether a structural break has occurred or not at a specified point in each time series by analyzing the time serie before and after. The dataset comprises tens of thousands of synthetic univariate time series, each containing approximately 2,000 to 5,000 values with a designated boundary point.
May 14, 2025
10:00 AM CETSeptember 15, 2025
4:59 PM CETEnd of October, 2025
| Name | Role | |
|---|---|---|
 | Dr. Horst D. Simon | Director, ADIA Lab |
 | Prof. Marcos Lopez de Prado | Global Head of Quantitative R&D at ADIA |
 | Prof. Alex Lipton | Global Head of Quantitative R&D at ADIA |
[] WEEK 1 — 12
In the first phase, participants are required to submit either a Python notebook (.ipynb) or Python script (.py) file. This file should contain the necessary code to build, load, or update their models and / or trained it on out-of-sample data. The code will be executed by the Crunch platform on a secure data old out. Participants can either use static models, trained only once on the initial training set, or dynamic models that update or retrain themselves on the unseen data, as explained further in the documentation.
[] WEEK 15
Crunchers can compete for a share of $100,000 in this structural break challenge. The top entry wins $40,000, with prizes awarded to the ten best submissions. The competition seeks to advance the field through creative problem-solving. More importantly, it's an opportunity for participants to showcase their skills and unique analytical solutions, regardless of demographics or academic background. The challenge welcomes innovative approaches to structural break from all corners.
| Position | Prize |
|---|---|
| 1 | $40,000 |
| 2 | $20,000 |
| 3 | $10,000 |
| 4 | $5,000 |
| 5 | $5,000 |
| 6 | $5,000 |
| 7 | $5,000 |
| 8 | $3,500 |
| 9 | $3,500 |
| 10 | $3,000 |
ADIA Lab is an independent Abu Dhabi-based institution engaged in basic and applied research in Data Science, Artificial Intelligence, Machine Learning, and High-Performance and Quantum Computing, across all major fields of study. This includes exploring applications in areas such as climate change and energy transition, blockchain technology, financial inclusion and investing, decision making, automation, cybersecurity, health sciences, education, telecommunications, and space. For more information, please visit www.adialab.ae or contact us at info@adialab.ae.