This talk reflects on the gap between calculated risk and experienced uncertainty – and on the uneasy task of investing when probability offers comfort, but not protection.
Machine learning (ML) models are highly flexible but face fundamental limits when learning from finite data. We show that a universal lower bound—the Limits-to-Learning Gap (LLG)—captures the unavoidable gap between the true (population) benchmark and linear ML models.
Henry Neville, a portfolio manager at MAN proposes the motion for the LQG annual debate : “Valuation is a paradigm that guarantees failure – this time it really is different…”
Optimizing portfolios using normalised returns delivers a factor-like awareness of underlying exposures while addressing the problem of sensitivity to model inputs with relative risk tolerance.
The LQG annual debate is the climax of the London Quant Group 2023 seminar series. Fun, informative and educational – Not to be missed
The legendary LQG autumn seminar – 11 awesome speakers will deliver seminars on quantitative investment to inform, inspire and discuss.
LQG 31/08/22 to 02/09/22 In-Person Autumn Seminar at Girton College, Cambridge with 11 speakers of international renown – after two years of lockdown it is great to be back!