Event Description
Traditional asset allocation frameworks require the estimation of expected returns and covariance matrices for constructing multi-asset portfolios. In practice, these estimations pose challenges that make them unreliable for optimization. In this paper, we introduce a systematic risk-based asset allocation framework which bypasses explicit returns forecast.
For the estimation of the covariance matrix, we employ Hierarchical Group Lasso (HGL) using a factor risk model with enhanced stability imposed by the sparsity on factor loadings. To refine the construction of the tactical asset allocation, we then introduce price-based signals, including momentum and low beta for traditional investments. For alternative investments, such as private assets and hedge funds, which typically have less frequent return observations and exhibit heavy-tailed return distributions, we incorporate their specific alphas to account for their systematic and idiosyncratic risks. These enhancements dynamically adjust exposures in response to market conditions without using return forecasts.
Our methodology bridges the gap between strategic asset allocation (SAA) and tactical asset allocation (TAA) decisions, offering a scalable and adaptable solution for institutional portfolios. By prioritizing a risk-based optimization and dynamic tactical adjustments, our framework enhances robustness and flexibility in asset allocation.
Event Description
Large Language Models (LLMs) are increasingly being integrated with reinforcement learning (RL) to push the boundaries of generalist AI agents. In finance, where real-time decision-making is critical, test-time compute efficiency plays a pivotal role in ensuring models can adapt dynamically to evolving market conditions. In-context reinforcement learning (ICRL) is emerging as a transformative approach, enabling LLMs to learn and refine on the fly without explicit fine-tuning. ICRL enhances adaptability in trading, risk assessment, and portfolio optimization. This paradigm shift moves us closer to AI agents capable of robust decision-making, paving the way for more autonomous and generalizable systems in high-stakes applications.
Event Description
This talk will cover several career pathways, essential skills, and recruitment trends specific to India. Plus, you’ll discover how the CQF program can uniquely position you to thrive in this competitive industry.
Why attend this event
- Gain a comprehensive understanding of career pathways in quantitative finance within India, highlighting the roles and responsibilities of quant traders, data scientists, risk analysts and portfolio managers.
- Find out how the CQF program can enhance your career prospects by providing cutting-edge quant finance and data science skills through a flexible, globally recognized qualification.
- Learn about the essential skills required to succeed in quant finance, including programming, mathematical modeling, and machine learning techniques.
- Explore insights into the recruitment landscape in India’s quant finance sector, preparing you for successful job applications and interviews.
Don’t miss this exclusive opportunity – book your free ticket today!
Event Description
This talk illustrates the enormous model risk that is present in the quantitative finance field and other domains. Various models calibrated to the same data can lead to significantly different results. Even within a single model, particular choices, like the objective function of a calibration, the numerical scheme of a Monte-Carlo simulation, the instruments to include or exclude from the calibration exercises can again lead to a variety of different outcomes. As a result, we must conclude that there is quite a bit of uncertainty around various pricing exercises. This issue is also present in other domains of science, like climate modelling, and hence one has to be cautious by using the outcome of a single model for policy making. Finally, we connect model risk with conic finance.
