Backtesting with correlation - A simple and effective pre-processing strategy to handle correlations
Wednesday 05 February 2025 | 13:00 - 13:50 | Webinar
This event has been organised by the the Data Science Working Group
Join us and unlock the secrets of effective backtesting. Dive into the world of time series datasets, where correlation and autocorrelation often complicate analysis. Learn how to leverage previous data points to enhance your strategies.
In this webinar Nikolai Nowaczyk, Risk management and AI consultant will give practical insights into backtesting strategies for counterparty credit risk. Nikolai will guide you through how to tackle challenges like data quality and computational intensity, and master techniques to handle correlation with ease.
Key learnings from this webinar will be:
- Understanding the impact of correlated samples on backtesting financial models and standard statistical tests.
- Exploring a simple pre-processing technique to decorrelate samples, improving test compatibility and discriminatory power.
- Seeing numerical examples that demonstrate more stable distributions, higher discriminatory power, and the ability to backtest multiple correlated quantities within a consistent framework.
Timings
Registration: 12:55
Event: 13:00 - 13:40
Nikolai Nowaczyk, Risk management and AI consultant
Nikolai Nowaczyk is a Risk Management & AI consultant who has advised more than 10 clients in over 20 projects around counterparty credit risk, xVA and model validation.
Nikolai is broadly interested in classical methods of financial mathematics and statistics as well as data science and machine learning.
Nikolai holds a PhD in mathematics from the University of Regensburg, has been an academic visitor to Imperial College London and likes to build bridges between academic research and practical applications.
Click here to view our event terms and conditions.