Quant Capital

🚀 Excited to share some novel research from AlgoLabs we've been working on with one of my former PhD students, W.J. Wisniewski! Check out our latest paper on innovative online clustering techniques to enhance portfolio allocation for traders here: https://bit.ly/ClusterAA, and see a static demo of our Generative AI models in action here: https://tabsoft.co/3Rrggea. 📄 Research Highlights: * We examined the real-world trades of 20,000 FX traders and discovered that clustering these trades significantly improves portfolio returns. * The study leverages the Ewens’ Sampling Distribution and Statistically Validated Networks (SVN) to better understand the temporal distributions of trader clusters. * Results show that AA, when applied to clustered trader data, struggles with too many similar trades but performs significantly better when clusters are utilised to manage this complexity. 🔗 Read the full paper here: https://bit.ly/ClusterAA Since publishing, we've extended our work to incorporate Generative AI techniques. By building a large space of automated trading models across various cross-asset symbols, we've created a diverse pool of trading models. These models can be integrated into our online clustering/AA framework to further stabilise returns. A follow-up paper and a live demo using real-time pricing data are on the way! 📈 Next Steps: * Using Generative Adversarial Networks (GANN), we generate trading models based on macro market conditions. * These models are then refined and validated using a discriminator network to ensure profitability. * The refined models are clustered and fed into the AA technique, showcasing improved stability and returns in live trading environments - we have a static demo here: https://tabsoft.co/3Rrggea, stay tuned for more updates and hopefully a live demo in the near future. Message me if you would like to find out more about this research! 🚀💡 #Research #Finance #Trading #MachineLearning #GenerativeAI #Clustering #PortfolioManagement