Quantitative Research

Models and tools for risk, forecasting, derivatives and much more

Quantitative Research applies advanced mathematical and statistical methods to open problems at the frontier of financial markets. The division develops models for risk measurement, forecasting and derivatives pricing, drawing on probability theory, statistical inference, optimisation and machine learning. Since the most interesting problems sit at the intersection of disciplines, the team deliberately brings together students from different quantitative backgrounds. Mathematics, economics, finance, AI and computer science all have a seat at the table: each member brings their own expertise and way of framing problems, and builds new ones by working alongside people who were trained entirely differently, replicating the collaborative structure of a real quantitative research team.

Meet the Team

How we work

Standards

Every paper states its research question, develops the theory with its derivations, describes the data with sources, sample periods and the treatment of missing values, and sets out the method. Results come with their validation, such as out-of-sample tests, bootstrap or heteroskedasticity-robust standard errors, residual diagnostics, convergence checks and repricing tests, followed by an explicit section on limitations, a conclusion and numbered references to the primary literature.

Review

Papers are written by small teams of analysts and are, in most cases, co-authored by the Head of Quantitative Research. Joint papers with other divisions, such as Macro Research, name every contributor. Every paper carries an academic disclaimer.

Tools and data

Python, with libraries such as scikit-learn, Keras, PyTorch and iisignature; Monte Carlo, PDE and Fourier methods; data from FRED, Yahoo Finance, FactSet, Robert Shiller's dataset and published regulatory and market datasets.

Latest Research Publications

Browse the full archive