Jiajing Sun

Jiajing Sun

PhD, CFA charterholder, FIMA, Member of the London Mathematical Society. Associate Professor and the Deputy Director of the Department of Statistics and Data Science, School of Economics and Management, University of the Chinese Academy of Sciences.

AI in Digital Finance — Online SeminarsWhen Does a Curve Really Change?

1 minute read

Published:

This post summarizes an online seminar from the AI in Digital Finance — Online Seminars series on how to detect economically meaningful shifts in functional time series, rather than changes that are only statistically significant.

Seminar Details

Speaker: Jiajing Sun (University of Chinese Academy of Sciences)
Date and time: March 6, 2026 (Friday), 11:00 (EEST) / 10:00 (CET)
Location: Online (Zoom)

Overview

In functional time series, standard change-detection tests can be too sensitive: they often identify shifts that are statistically significant, but too small to be relevant in real decisions. This talk presents a self-normalized, adjusted-range testing approach that asks a more practical question: is the change in the curve large enough to matter?

The method works directly on functional observations and sidesteps common tuning choices (e.g., bandwidth selection) as well as routine dimension-reduction steps such as FPCA. The seminar also covers large-sample validity and a simulation-based calibration for the classical “no-change” benchmark.

To illustrate the approach, we study executed trade-level Bitcoin options from Deribit, build daily 30-day constant-maturity implied-volatility smiles on a shared log-moneyness grid, and run a rolling-window monitor. With a relevance threshold of one volatility point (10% level), the monitor flags a small set of economically meaningful regime changes (notably around early October and early December 2025). A simple straddle example shows how relevance-gated signals can be translated into low-turnover decision rules.

Zoom link: Join the seminar
Event page: AI in Digital Finance seminar series

Key Takeaways

  • How to distinguish statistical detectability from practical relevance in curve changes.
  • A tuning-light procedure for monitoring shifts in functional time series.
  • A worked crypto-derivatives case study using implied-volatility smiles.

Jiajing Sun’s research spans digital finance and data-driven methods, including work on building multi-dimensional evaluation frameworks for financial technology competitiveness using bank annual-report data and machine-learning/text-analysis pipelines.

You May Also Enjoy