Kritzman's research, conducted along with Huili Song, quantitative researcher at State Street and David Turkington, head of State Street Associates, introduced the concept of dynamic time warping, a technique originally used in speech recognition, to synchronize the evolution of stock price bubbles that often follow irregular patterns.

Huili Song is a Vice President and Quantitative Researcher at State Street Associates (SSA). Since joining SSA in 2020, Huili has worked on quantitative applications of SSA’s research and led product development initiatives for State Street Global Markets’® research platform - Insights.

Nov 19, 2024 · Unlike narrow statistical models, large language models (LLMs) can extrapolate patterns across disparate domains of knowledge. We apply a logical interpretation framework called Model Fingerprint to quantify the precise macroeconomic reasoning of LLMs.

Dec 16, 2023 · Kritzman, Mark and Song, Huili and Turkington, David, Dynamic Warp Analysis: A New Approach for Detecting and Timing Bubbles (December 12, 2023). MIT Sloan Research Paper No. 6960-23, Available at SSRN: https://ssrn.com/abstract=4664898 or http://dx.doi.org/10.2139/ssrn.4664898.

Sep 8, 2022 · We propose a parsimonious yet flexible statistical method for predicting the relative vulnerability or resilience of individual stocks to market drawdowns.

Affiliations: [College of Intelligent Systems Science and Engineering, Harbin Engineering University, Harbin, China].

Semantic Scholar profile for Huili Song, with 1 highly influential citations and 10 scientific research papers.

Nov 18, 2024 · By Megan Czasonis, Huili Song, and David Turkington. We show that LLMs can effectively extrapolate from disparate domains of knowledge to reason through economic relationships, and that this may have advantages over narrower statistical models. Fundamentally, large language models (LLMs) and numerical models both learn patterns in training data.