Month: April 2026

Volatility Risk Premium and Clustering: Intraday vs Overnight Dynamics

The decomposition of risks and returns into overnight and intraday components is an emerging area of research. In this post, we examine how these components differ in terms of volatility clustering and the variance risk premium, and what this implies for forecasting, risk management, and strategy design.

Breaking Down the Volatility Risk Premium: Overnight vs. Intraday Returns

The decomposition of the volatility risk premium (VRP) into overnight and intraday components is an emerging area of research. Most studies indicate that the VRP serves as compensation for investors bearing overnight risks.

Reference [1] continues this line of research, with its main contribution being the decomposition of the variance risk premium into overnight and intraday components using a variance swap approach. The study also tests the predictive ability of these components and examines the seasonality (day-of-week effects) of the VRP.

Findings

-The paper decomposes the variance risk premium into overnight and intraday components across the US, Europe, and Asia.

-It finds that the variance risk premium is significantly negative during the overnight non-trading period.

-During the intraday trading period, the variance risk premium becomes positive and often insignificant.

-The results show that the overall negative variance risk premium documented in prior studies is largely driven by the overnight component.

-The study uses the P&L of a hypothetical variance swap to analyze these components.

-The intraday variance risk premium captures short-term risk and has predictive power over 1 to 3-month horizons.

-The overnight variance risk premium reflects longer-term risk and shows predictive ability over 6 to 12-month horizons.

-The findings highlight the importance of non-trading periods in explaining the behavior of the variance risk premium.

In summary, the study reaffirms that the variance risk premium is significantly negative during the non-trading overnight period, while it becomes positive and often insignificant during the intraday trading period.

An interesting finding is the day-of-week seasonality. For instance, going long volatility at the open and closing the position at the close tends to be profitable on most days, except Fridays.

Reference

[1] Papagelis, Lucas and Dotsis, George, The Variance Risk Premium Over Trading and Non-Trading Periods (2024). SSRN 4954623

Intraday and Overnight Volatility Clustering Effect

Volatility clustering is a phenomenon observed in financial markets where periods of high volatility tend to cluster together, followed by periods of low volatility. This pattern suggests that extreme price movements are not randomly distributed over time but rather occur in clusters or groups.

Volatility clustering has undergone extensive study within the daily timeframe. Reference [2] delves into volatility clustering within intraday and overnight timeframes. It specifically investigates clustering within each timeframe and between them.

Findings

-The paper studies volatility clustering in intraday and overnight returns across 15 global equity markets.

-It finds that volatility clustering is present in both intraday and overnight returns across multiple time scales, from daily to long-term horizons.

-The results show that volatility clustering is generally stronger in overnight returns than in intraday returns.

-Cross clustering between intraday and overnight volatility is relatively weak within each market, especially at shorter time scales.

-The findings are consistent across both developed and emerging markets, indicating a universal pattern.

-The study highlights the importance of considering both short-term and long-term risks in equity markets.

-The results suggest that volatility dynamics differ between trading and non-trading periods.

-The paper provides implications for trading and risk management strategies based on volatility clustering behavior.

In short, the paper shows that volatility clustering is a universal feature of both intraday and overnight returns across multiple time scales. It also finds that clustering is stronger overnight, while cross-effects between intraday and overnight volatility remain weak, with consistent patterns across global markets.

Reference

[2] Xiaojun Zhao, Na Zhang, Yali Zhang, Chao Xu, Pengjian Shang, Equity markets volatility clustering: A multiscale analysis of intraday and overnight returns, Journal of Empirical Finance 77 (2024) 101487

Closing Thoughts

Taken together, these studies show that volatility dynamics differ significantly between intraday and overnight periods, both in terms of risk pricing and clustering behavior. The variance risk premium is largely driven by the overnight component, while intraday and overnight volatility exhibit distinct clustering patterns with limited interaction. These findings highlight the importance of separating trading and non-trading periods in both forecasting and risk management, as each captures different horizons and sources of risk, offering more refined inputs for portfolio construction and strategy design.

Large Language Models in Trading: Models and Market Dynamics

I just returned from a two-day conference in New York, FutureAlpha (formerly QuantStrats). This year, the theme focused largely on data, machine learning, and AI. While some speakers were very enthusiastic about the potential of AI to generate alpha, our panel was more conservative. The consensus among the panelists was to use ML and AI to enhance and improve risk management. Along this theme, in this post, I discuss the use of generative AI in trading.

Integrating Structured and Unstructured Data with LLMs and RAG

Traditional quantitative methods often rely on structured data, such as time series. With the emergence of Large Language Models (LLMs), it is now possible to process unstructured data. A new line of research focuses on integrating unstructured data analysis into traditional frameworks.

Along this line, Reference [1] proposed the use of LLMs together with retrieval-augmented generation (RAG) to process both structured and unstructured data concurrently. Specifically, the authors developed a system that first applies LLMs to detect regime shifts using time-series techniques, then employs RAG to integrate external knowledge into the model’s decision-making process. By retrieving relevant information from a vector database and combining it with the model’s capabilities, RAG improves both the interpretability and effectiveness of trading strategies.

Findings

-The paper studies methods for fine-tuning open-source Large Language Models to enhance quantitative trading strategies.

-It integrates numerical data, such as prices and technical indicators, with textual data, including news and sentiment.

-The approach uses Retrieval-Augmented Generation with a vector database to process and contextualize textual information.

-The study focuses on fully fine-tuning smaller models to achieve cost efficiency and scalability.

-It proposes a hybrid framework that combines LLM capabilities with traditional quantitative methods.

-The framework incorporates real-time data pipelines and adaptive model tuning.

-The results show improvements in predictive accuracy and risk-adjusted returns.

-The integration of multimodal data helps address challenges in combining structured and unstructured information.

-Fine-tuned smaller models improve regime detection and trading decision accuracy while maintaining efficiency.

-Additional techniques enhance model performance and robustness, supporting practical applications in quantitative finance.

In short, incorporating RAG into the framework enhances the model’s ability to understand complex macroeconomic environments and adapt trading strategies as conditions evolve. Experimental results show significant gains in predictive accuracy and risk-adjusted returns, demonstrating the practical value of these fine-tuning methods in finance.

Reference

[1] Li, C., Chan, C.H.R., Huang, S.H., Choi, P.M.S. (2025). Integrating LLM-Based Time Series and Regime Detection with RAG for Adaptive Trading Strategies and Portfolio Management. In: Choi, P.M.S., Huang, S.H. (eds) Finance and Large Language Models. Blockchain Technologies. Springer, Singapore.

Can AI Trade? Modeling Investors with Large Language Models

The previous paper focuses on improving trading performance by integrating LLMs with quantitative models and data, while another line of research explores how LLMs behave as autonomous agents within market environments.

Reference [2] utilized LLMs to construct trading agents in the financial markets. Specifically, the author used LLMs to emulate various types of investors: value investors, momentum traders, market makers, retail traders, etc.

Findings

-The paper develops a simulated stock market in which large language models act as heterogeneous trading agents.

-The framework includes realistic market features such as an order book, market and limit orders, partial fills, dividends, and equilibrium clearing.

-Agents operate with different strategies, information sets, and endowments, and communicate decisions using structured outputs while explaining reasoning in natural language.

-The results show that LLMs can consistently follow instructions and implement strategies such as value investing, momentum trading, and market making.

-LLM agents process market information and respond meaningfully to prices, dividends, and historical data.

-The simulated market exhibits realistic dynamics, including price discovery, bubbles, underreaction, and liquidity provision.

-The framework enables controlled analysis of agent behavior under different market conditions, similar to interpretability methods in machine learning.

-It provides a cost-effective way to test financial theories that lack closed-form solutions.

-The study highlights that LLM behavior is highly sensitive to prompts, which can lead to correlated actions across agents.

-This correlation may amplify volatility and introduce systemic risks, emphasizing the need for careful testing before real-world deployment.

In short, the article concluded that trading strategies generated by large language models are effective, but could introduce new systemic risks to financial markets because these agents would act in a correlated manner.

Reference

[2] Alejandro Lopez-Lira, Can Large Language Models Trade? Testing Financial Theories with LLM Agents in Market Simulations, arXiv:2504.10789

Closing Thoughts

In this issue, the discussion highlights two complementary directions in applying LLMs to finance. On one hand, integrating LLMs with quantitative models and multimodal data can improve predictive accuracy and risk-adjusted returns. On the other hand, treating LLMs as autonomous trading agents reveals how their behavior can shape market dynamics, including liquidity, price discovery, and potential instability. Taken together, the results suggest that while LLMs offer meaningful opportunities in trading and risk management, their impact depends critically on implementation, prompting, and control of system-wide behavior.