Forecasting duration in high-frequency financial data using a self-exciting flexible residual point process

Forecasting duration in high-frequency financial data using a self-exciting flexible residual point process

Swiss Finance Experts Tackle Complexities of High-Frequency Trading with Innovative Method

Section 1 – What happened?

Researchers at the Swiss Finance Institute have developed a novel approach to forecasting durations in high-frequency financial data, specifically in limit order books. The method, based on a self-exciting flexible residual point process, aims to tackle the challenges posed by heavy-tailed interarrival times in modern exchanges. According to the study, the proposed approach outperforms several alternative methods in forecasting durations in ultra-high-frequency trading data.

Section 2 – Background & Context

High-frequency trading has become increasingly prevalent in modern financial markets, with exchanges processing millions of trades per second. However, the complexities of high-frequency data pose significant challenges for accurate prediction and risk management. Traditional forecasting methods often struggle to capture the intricate patterns and dependencies in high-frequency events, leading to suboptimal results. The Swiss Finance Institute's research addresses this issue by developing a novel method that incorporates empirical distributional features and preserves the self-exciting and decay structure of the data.

Section 3 – Impact on Swiss SMEs & Finance

The innovative forecasting method developed by the Swiss Finance Institute has significant implications for Swiss SMEs and the broader financial sector. By providing more accurate predictions of high-frequency trading events, the method can help financial institutions and traders make more informed decisions, reducing the risk of losses and improving overall market efficiency. Additionally, the method can be applied to various financial instruments and markets, making it a valuable tool for risk management and portfolio optimization.

Section 4 – What to Watch

The Swiss Finance Institute's research has opened up new avenues for exploring the complexities of high-frequency financial data. As the method continues to be refined and applied to real-world scenarios, it will be essential to monitor its performance and adaptability in different market conditions. Furthermore, the potential applications of the method extend beyond high-frequency trading, making it a promising area of research for the broader finance community.

Source

Original Article: Forecasting duration in high-frequency financial data using a self-exciting flexible residual point process

Published: April 1, 2026

Author: Kyungsub Lee

---

Disclaimer: This article is for informational purposes only and does not constitute financial advice. Consult a licensed financial advisor before making investment decisions.