AI Hype vs Physics Envy: The Quant Conundrum 

There is an increasing temptation for quants to use AI for various applications in quantitative finance, from identifying statistical arbitrage to predicting crashes, hedging and fitting volatility surfaces. We begin by comparing the different pros and cons of more parameterised AI based models to statistics-based and econophysics-based ones.

 

We take crash prediction as an example and compare the robustness of a standard hundred parameter LSTM, to the six parameter econophysics based LPPL model, to finally physics-inspired three and one parameter models. We see that at each level of parsimony, there is an increase in robustness and explainability, and decrease in false signals. We end with how AI can compete with or incorporate physics based modelling in low signal to noise regimes going forward.

Revant Nayar is currently Principal and CTO at FMI Tech, among the first quantum AI based quantitative hedge fund and Research Affiliate at Stony Brook University. He started his career as a theoretical physics researcher at IAS and Princeton University with a specialty in quantum field theory and string theory, where he started FMI Tech as a think-tank applying cutting edge innovations in quantum and classical physics to data science and quantitative finance. Subsequently, he served as Research Fellow at NYU Tandon where he developed the first ever group theoretic method to compute options pricing density functions with Peter Carr. 

 

In 2021, he became among the first fund managers applying quantum alternatives to AI to asset management starting with the Indian market and recently expanding to the US market. He has over the years been featured as speaker and panellist at the leading finance and AI conferences including Bloomberg, Strata, Global AI Conference, YPO, NYU and EMEX. He also runs among the largest econophysics research collaborations in the world (FTERC), bringing students, researchers and professors from top universities around the world to generate research at the intersection of physics, mathematics and quantitative finance. Today he is widely regarded as a trailblazer in quantum and econophysics-inspired approaches to asset and risk management and among the promising emerging quant fund managers.