Arkin Gupta

This talk dives into the evolving landscape of equities risk modeling, blending tried-and-true fundamentals with the latest advancements in data science. We’ll start by covering the basics: the role of risk models, sample covariance estimation, and techniques like covariance matrix shrinkage. From there, we’ll explore multi-factor fundamental models and statistical risk models, laying the foundation for a comparison with cutting-edge deep learning approaches. Throughout the session, we’ll tackle practical questions like: How do traditional models stack up against deep learning in today’s markets? When is interpretability more important than predictive power? And how can we overcome the computational and data challenges these methods present? This session is designed for finance and data science professionals who want to stay ahead of the curve. We’ll wrap up with a Q&A, giving participants the chance to engage directly and explore how these concepts apply to their own work.