Siddartha Ghoshal

DPhil Student, Engineering, Machine Learning Research Group
Sid Ghoshal graduated from Imperial College with a BSc in Mathematics and Management. Following a year as an analyst in corporate finance at Dresdner Kleinwort Wasserstein, he completed an MSc in Finance and Economics at the LSE with a focus on portfolio management, exploring the role of momentum in the seminal Fama-French 3-factor model. Returning to industry as a derivatives trader for Deutsche Bank's Complex Risk Group, he traded their commodities correlation exotics book for 5 years, eventually lead-managing their risk during DB's ascent to IFR Commodity Derivatives House of the Year in 2008.
Sid returned to academia for fresh insights in systematic trading, earning his MSc in Computer Science at Oxford in 2012 with thesis work on ternary commodity return classification using Support Vector Machines, assessing both the uneven stochasticity of the business cycle and the need for parsing newsflow on geopolitically sensitive assets like crude oil. 
Supported for doctoral research by the EPSRC and ESRC, he joined the Autonomous, Intelligent Machines and Systems CDT led by Prof. Steve Roberts in 2014, with a view to developing state-of-the-art Bayesian techniques for fusing heterogeneous data streams in time series forecasting.

Related Events

10th Anniversary Oxford-Man Institute Annual Workshop

Working Paper

Bengtzen, M., Ghoshal, S. and Roberts, S (2016). Pre-earnings announcement drift: Inferring informed trading from the tape.

Published Research

Ghoshal, S. and Roberts, S. (2016). Extracting predictive information from heterogeneous data streams using Gaussian processes. Algorithmic Finance. 5(1-2). 21-30.