Applied Physics/Physics Colloquium: Shamit Kachru- "From Random Walks to Phase Transitions: Information and Trading"

Abstract: I will discuss ideas in modern quantitative finance from the perspective of a theoretical physicist. Starting from the classic picture of price fluctuations as random walks, I will explain how practitioners identify collective modes, or “factors,” and use them to isolate residual predictability, or “alpha.” The problem of trading on such predictions naturally becomes a problem in control theory, with close analogies to signal transmission through a noisy channel. In a simple model, one finds a cascade of bifurcation transitions: from no trading, to discrete buy/hold/sell policies, to progressively finer discretizations, with continuous linear-quadratic control recovered in an appropriate limit.


Shamit Kachru is Professor Emeritus of Physics at Stanford University, where he served on the faculty for 25 years, including terms as Chair of the Physics Department and Director of the Stanford Institute for Theoretical Physics. His research in theoretical physics focused on string theory, quantum field theory, and applications to cosmology and condensed matter physics. His honors include a Sloan Fellowship, a Packard Fellowship, a Simons Investigator Award, and election to the American Academy of Arts and Sciences. Since 2023, he has been an Executive Director at PDT Partners, a quantitative investment firm, where his research applies ideas from theoretical physics, statistical learning, and control theory to problems in systematic trading.