Tom Spencer

Data scientist, econometrician, quantitative researcher and statistician with expertise in novel econometric and machine learning approaches to dynamic factor models, risk modelling, and quantitative analysis across investment banking and asset management sectors.

Internship, AI engineering at Quotes and Plans ltd (2026-03 – Present)

AI powered construction quotes and plans app. Developing a multimodal GUI which dynamically fills out forms using a speech-to-speech chatbot with python and Google Gemma AI / Hugging Face backend.

Senior Model Validator (manager level) at ICBC Standard Bank (2017-04 – 2021-07)

Part of a small new team responsible for validation of all of the Bank's mathematical modelling techniques (mainly risk and valuation), model risk management, model risk framework and governance.

• Risk models: IRC (PD / LGD credit risk predictive model using boundary separation algorithms and cross-validated decision tree-based scorecard models, data pipelines etc.), IRRBB (loans and deposits behavioural repayment model), Value at Risk (VaR) model (GARCH, analysis of statistical bias due to overlapping historical data shifts, appropriateness of arithmetic / geometric risk factor shifts), various Risk Not In VaR (RNIV) models, hypothesis testing, backtesting and stress tests.
• Pricing / NPV models: Included loans, bonds, interest rate swaps, CDS, swaptions and energy derivatives.
• Counterparty credit risk pricing models (XVA): Included Monte Carlo interest rate simulation model for Credit Value Adjustment and spread-adjusted discounting curve model for Funding Value Adjustment.

Quantitative Analyst at HSBC (2015-09 – 2017-03)

Counterparty Credit Risk calculation, analysis & development for the investment bank globally.

• Main analyst for the transition from the Black-Karasinski interest rates model (calibrated econometrically using macroeconomic inputs) to the new market implied Hull-White Monte Carlo simulation model. Included validation of the models, comparison to analytical approximations.
• Originated a novel P&L statistical attribution analysis procedure based on fitting Gram-Charlier distributions to the simulated future portfolio Mark to Market.

Senior Business Consultant at Avaloq Evolution AG (2012-04 – 2015-09)

Development of the Avaloq Banking System to meet asset management client requirements, specialising in risk and valuation models. Often working at client sites, mainly Barclays wealth and Canaccord wealth.

• Implementation and validation of the asset valuation engine to ensure correct pricing and risk sensitivities calculation for derivatives, bonds, FX and interest rate products.
• Implementation and validation of portfolio construction approaches, hedging, risk (VaR, P&L attribution, stress scenarios), performance attribution and regression based risk factor sensitivity models.

Quantitative Analyst at Otkritie Securities (2011-05 – 2011-11)

A key part of new research team within the investment bank aimed at providing innovative quant and econometric research and tools to clients and our traders.

• Dynamic Factor Model to forecast interest rates and corporate bond spreads.
• Probit & logistic regression models to estimate the probabilities of a recession in the US and Eurozone
• Strategies for pairs trading using cointegration, bond relative value analysis, intrinsic value models for distressed debt products, Asian energy warrants and structural & reduced form credit risk models.

Performance Analyst at Northern Trust (2008-02 – 2011-05)

Performance, risk and attribution analysis across asset classes (equities, fixed income, FX, IR, options etc).

PhD in Economics – Birkbeck, University of London (2019 – 2025)

Course in Machine Learning – Stanford University (2018 – 2019)

MSc in Financial Engineering – Birkbeck, University of London (2007 – 2009)

BSc in Mathematics – King's College, University of London (2002 – 2006)

A Level / A/S Level in Mathematics, Economics, Physics, Further Mathematics, Music – Cardinal Vaughan Memorial School (1995 – 2002)