World Business StrategiesServing the Global Financial Community since 2000

Main Conference Day 2: Friday 26th September

08.30 – 09.00: Morning Welcome Coffee

Morning Stream Chair:

Ioana Boier:

AI / LLMs / ML Stream

09.00 – 09.45: Accelerated Portfolio Optimization and AI methods

Ioana Boier:

AI / LLMs / ML Stream

09.45 – 10.30: ML Portfolios vs. the S&P 500: Forecasting Returns

  • Gradient Boosting vs. LSTM in forecasting S&P 500 returns.
  • Does dimensionality reduction improve model learning?
  • Rethinking performance measures.
  • Practical pitfalls in model selection and implementation.

We revisit the problem of predicting stock returns for S&P 500 constituents using LSTM and Gradient Boosting, comparing not only their out-of-sample predictive accuracy, but also the portfolio returns implied by these forecasts. Traditional performance metrics like RMSE dominate the evaluation of forecasting models, though they can be misleading when the true objective is financial performance. Surprisingly, our results show that Gradient Boosting significantly outperforms LSTM in realized portfolio returns, even in periods when RMSE suggests the opposite.

Miljana Cvetkovic:

Independent researcher

10.30 – 11.00: Morning Break and Networking Opportunities

AI / LLMs / ML Stream

11.00 – 11.45: Foundational Models for Financial Time-Series Analysis

Arun Verma:

Head of Quantitative Research Solutions, Bloomberg
AI / LLMs / ML Stream

11.45 – 12.30: Garbage In, Garbage Out: The Critical Role of Training Data in AI

Christopher Kantos:

Managing Director and Head of Quantitative Research, Alexandria Technology

12.30 – 13.30: Lunch

Afternoon Stream Chair: TBC

AI / LLMs / ML Stream

13.30 – 14.15: Model Risk Management for AI/ML models

With the help of use cases, we discuss the need and adoption of new techniques and testing framework for model risk management of AI/ML models.

Harsh Prasad:

Principal and CEO, Qxplain
AI / LLMs / ML Stream

14.15 – 15.00: NeuralFactors: A Novel Factor Learning Approach to Generative Modeling of Equities

Presenter to be confirmed.

15.00 – 15.30: Afternoon Break and Networking Opportunities

AI / LLMs / ML Stream

15.30 – 16.15: ML technique applied to Sport trading

Maurizio Garro:

Senior Lead – IBOR Transition programme, Lloyds Banking Group

08.30 – 09.00: Morning Welcome Coffee

Morning Stream Chair: TBC

Volatility / Options / Monte Carlo Stream

09.00 – 09.45: “The unreasonable effectiveness of Randomized Quasi-Monte Carlo in option pricing and risk analysis”

We compare three approaches:

  • Standard Monte Carlo (MC)
  • Quasi-Monte Carlo (QMC)
  • Randomized Quasi-Monte Carlo (RQMC)

While MC methods are widely used, they converge slowly. QMC improves convergence using low-discrepancy sequences, particularly for problems with low effective dimension. However, since QMC is deterministic, it lacks reliable error estimation. This is where RQMC comes in, using randomized sequences like Owen’s nested scrambling to allow for error estimation.

To further improve performance, we apply variance reduction techniques like Brownian Bridge (BB) and Principal Component Analysis (PCA). These reduce effective dimension and enhance convergence.

Our numerical tests on Asian options show that RQMC, especially when combined with PCA, significantly outperforms standard MC in both accuracy and efficiency. We also analyze convergence behavior for price and Greeks (Delta, Gamma), and find that RQMC with PCA consistently delivers the best results, particularly in high-dimensional scenarios.

Julien Hok:

Quantitative Analysis, Investec Bank

Sergei Kucherenko

Senior Research Fellow, Imperial College

Volatility / Options / Monte Carlo Stream

09.45 – 10.30: ‘A Hype- Adjusted Probability Measure for NLP Forecasting of Stock Price Volatility’

Helyette Geman:

Ralph O’Connors Sustainable Energy Institute, Johns Hopkins University

10.30 – 11.00: Morning Break and Networking Opportunities

Volatility / Options / Monte Carlo Stream

11.00 – 11.45: Stationary Portfolio Construction

Portfolio allocation is crucial for investment institutions with diverse market functions and strategies. Allocations can be static, based on long-term expectations, or dynamic, incorporating time-varying stochastic factors. A typical dynamic allocation goal consists of an expectation of a deterministic function of the portfolio wealth at a fixed time-horizon. Usually, the portfolio weights depend on prices, wealth, and a time to maturity.

Our paper explores a novel optimization framework without fixed time horizons, focusing on portfolio returns rather than wealth – an approach particularly relevant to sovereign and pension funds. We adopt a comprehensive stationary approach across utility (regularized running forward-looking returns), model (asset returns driven by stationary market factors like sentiments and VIX), and weights (functions of market drivers without explicit time-dependence).

We derived optimal equations for asset weights with arbitrary utility functions, potentially solvable numerically. Additionally, we developed an efficient convex optimization routine for a special case of concave utilities. Morever, for quadratic utilities, we found out that the optimal weights satisfy a matrix Fredholm equation and develop a highly efficient numerical solution. Finally, we obtained explicit analytics for the Kim-Omberg model solving the Fredholm equation. We have confirmed our findings via numerous numerical experiments: we calibrated the Kim-Omberg model using market sentiment as the stochastic drift factor and holistically analyzed analytical and numerical results.

Alexandre Antonov:

Quantitative Research & Development Lead,
Volatility / Options / Monte Carlo Stream

11.45 – 12.30: Volatility Parametrizations with Random Coefficients: Analytic Flexibility for Implied Volatility Surfaces

It is a market practice to express market-implied volatilities in some parametric form. The most popular parametrizations are based on or inspired by an underlying stochastic model, like the Heston model (SVI method) or the SABR model (SABR-parametrization). Their popularity is often driven by a closed-form representation enabling efficient calibration. However, these representations indirectly impose a model-specific volatility structure on observable market quotes. When the market’s volatility does not follow the parametric model regime, the calibration procedure will fail or lead to extreme parameters, indicating inconsistency. This article addresses this critical limitation – we propose an arbitrage-free framework for letting the parameters from the parametric implied volatility formula be random. The method enhances the existing parametrizations and enables a significant widening of the spectrum of permissible shapes of implied volatilities while preserving analyticity and, therefore, computation efficiency. We demonstrate the effectiveness of the novel method on real data from short-term index and equity options, where the standard parametrizations fail to capture market dynamics. Our results show that the proposed method is particularly powerful in modeling the implied volatility curves of short expiry options preceding an earnings announcement, when the risk-neutral probability density function exhibits a bimodal form.

Lech Grzelak:

Quantitative Analyst, Rabobank and Assistant Professor, TUDelft

12.30 – 13.30: Lunch

Afternoon Stream Chair: TBC

Volatility / Options / Monte Carlo Stream

13.30 – 14.15: Pricing and calibration in the 4-factor path-dependent volatility model (to be confirmed)

Julien Guyon: 

Full Professor, Ecole des Ponts ParisTech

Volatility / Options / Monte Carlo Stream

14.15 – 15.00: Some aspects of valuing autocallable notes

We look into challenges of producing accurate valuation and risk metrics for the autocallable notes. The focus is on the Monte Carlo methods, which cater for the richness of features and high dimensionality of the general product population. Our investigation focuses more closely at the application of the classic techniques, such as stratified sampling, as well as some more recent approaches based on signatures.

Vladimir Lucic

Head of Quants, Marex Solutions & Visiting Professor, Imperial College London

15.00 – 15.30: Afternoon Break and Networking Opportunities

Volatility / Options / Monte Carlo Stream

15.30 – 16.15: From Black-Scholes to Q-Learning: Explainable Reinforcement Learning for Robust Option Pricing and Hedging

  • QLBS model: where tractability meets performance.
  • Bridging theoretical framework and practical application.
  • Robustness across different market conditions.
  • Empirical analysis on S&P 500 options.

RL has become increasingly present in financial modelling, yet many models remain overly reliant on black-box approximations. We approach the problem of option pricing and hedging through the QLBS model – an explainable RL framework proposed by Halperin that integrates traditional replication logic with AI-based decision making. Our results reveal that QLBS achieves robust pricing and hedging performance across different states and market conditions. In addition, we incorporate proportional transaction costs into the model and evaluate the resulting P&L for S&P 500 options.

Zoran Stoiljkovic:

Independent researcher

08.30 – 09.00: Morning Welcome Coffee

Morning Stream Chair: TBC

Modelling / xVA / Regs Stream

09.00 – 09.45: Credit – Equity Revisited

Most counterparties do not have traded CDS instruments. This poses a challenge for calculating CVA which relies on risk-neutral default probabilities.

Here we present a new model for the estimation of credit spreads using equity market data.

In contrast to more traditional credit-equity models, we take an empirical approach in order to determine a simple functional relationship that can be used in practice for CVA risk management.

We find that our model out-performs models that rely solely on credit data as well as alternative credit-equity models in the literature.

Matthias Arnsdorf:

Global head of Counterparty Credit Risk Quantitative Research, J.P. Morgan

Modelling / xVA / Regs Stream

09.45 – 10.30: Topic to be confirmed.

Michael Pykhtin:

Manager, Quantitative Risk, U.S. Federal Reserve Board

10.30 – 11.00: Morning Break and Networking Opportunities

Modelling / xVA / Regs Stream

11.00 – 11.45: “The Pricing of XVA and Stochastic Corporate Liabilities”

Andrey Chirikhin:

Founder, Quantitative Recipes

Modelling / xVA / Regs Stream

11.45 – 12.30: Shock Propagation from Equity to Debt

* Using a unique database of millions of corporate bonds, we investigate how shocks to firms spread to their equity and debt, and ultimately to individual bonds.

* We show that shock to bonds is significantly higher at times of systemic stress, indicating behavioural effects of a heightened risk premium associated with corporate bonds.

* We also demonstrate how individual firm and bond characteristics influence this shock propagation.

* Applications of our results range from climate stress testing to geopolitical risks – we will explore these applications to diversified portfolio construction and risk management.

Svetlana Borovkova:

Climate Risk Quant Research, Bloomberg

12.30 – 13.30: Lunch

Afternoon Stream Chair: TBC

Modelling / xVA / Regs Stream

13.30 – 14.15: FRTB: Evolution of Regulatory Landscape and Implications

Adolfo Montoro:

Director, Global Market Risk Analytics, Bank of America

Modelling / xVA / Regs Stream

14.15 – 15.00: Potential Perils of Desk Level Backtesting under FRTB

Eduardo Epperlein:

Managing Director, Global Head of Risk Methodology, Nomura International

15.00 – 15.30: Afternoon Break and Networking Opportunities

  • Discount Structure
  • Super early bird discount
    20% until 6th June 2025

  • Early bird discount
    15% until 18th July 2025

  • Early bird discount
    10% until 15th August 2025

  • Special Offer
    When two colleagues attend the 3rd goes free!

  • 70% Academic Discount
    (FULL-TIME Students Only)

Download Event PDF

Please enter your details to download this PDF

Error

Event Email Reminder

Error