Machine learning
Advanced visualization for the quant strategy universe: clustering and dimensionality reduction
The authors present a novel visualisation model, based on 5000 quantitative investment strategies, which can identify nonlinear relationships and clustering strategies with similar risk factor exposures.
A model combining Optuna and the light gradient-boosting machine algorithm for credit default forecasting
The authors put forward a default prediction model designed to make the analysis of complex, highly dimensional and imbalanced real-world bank data easier.
Banks must loosen up on ChatGPT use – risk chiefs
Risk Live: ’Shadow use’ and inability to attract new hires mean restricting access to GPTs is untenable
Quantum cognition machine learning: financial forecasting
A new paradigm for training machine learning algorithms based on quantum cognition is presented
Dutch regulator in new push on algo manipulation
AFM teams up with Oxford Uni academics to develop data models that will identify “harmful” activity in automated trading
Analyzing credit risk model problems through natural language processing-based clustering and machine learning: insights from validation reports
The authors use clustering and machine learning techniques to analyze validation reports, providing insights to the development, implementation and maintenance of credit risk models.
Machine learning prediction of loss given default in government-sponsored enterprise residential mortgages
The authors apply machine learning techniques to Loss Given Default estimation, identifying key variables in LGD prediction and evaluating the performance of various models.
Forecasting India’s foreign trade dynamics: evaluation of alternative forecasting models in the post-pandemic period
The authors aim to determine how India's foreign trade will change following Covid-19 and the Russia-Ukraine conflict, comparing several forecasting models and identifying that which performs best.
Derivatives pricing with AI: faster, better, cheaper
Pascal Tremoureux, head of quantitative research at Murex, describes the firm’s mission to replicate derivatives pricing models through machine learning – slashing time and costs in the process
Clustering market regimes using the Wasserstein distance
The authors apply Wasserstein distance and barycenter to the k-means clustering algorithm, validating their proposed method both qualitatively and quantitatively.
Risk Technology Awards 2024: AI hopes and holdups
Live AI use-cases are limited, as vendors warn on over-regulation
Chartis RiskTech AI 50 2024
A Chartis report exploring the landscape of artificial intelligence and its adoption in the financial services
CVA sensitivities, hedging and risk
A probabilistic machine learning approach to CVA calculations is proposed
Podcast: Alvaro Cartea on collusion within trading algos
Oxford-Man Institute director worries ML-based trading could have anti-competitive effects
New data techniques to turbocharge risk management
Risk management and data management have become central to the broader digitalisation efforts of financial institutions. A robust data strategy has an important role in supporting and enhancing risk management efforts
Quants see promise in DeBerta’s untangled reading
Improved language models are able to grasp context better
An equity-implied rating model for unrated firms
The authors use Merton's distance to default as the basis for new model with which to assign credit ratings to firms which are not traditionally rated.
FX automation plans focus on predictive analytics – panel
Panellists suggest banks could explore AI tools for foreign exchange pricing
The coming AI revolution in QIS
The first machine learning-based equity indexes launched in 2019. They are finally gaining traction with investors
Hedge fund’s bots hunt for ‘non-linear’ trade signals
Boutique investment firm Goose Hollow uses LLMs to scrape thousands of news sources, searching for links that others miss
Banks, vendors mine AI for corporate FX hedging
New machine learning algos can help corporate clients adjust hedging ratios, but tech’s effectiveness is limited by data quality, experts caution