Candlestick-Based Spot Correlation Estimation
Prof. Qiyuan LI
Assistant Professor in Economics
HKU Business School
University of Hong Kong
Date: 7 May 2026 (Thursday)
Time: 09:30-11:00
Venue: E22-G008
Host: Prof. Jun YU, UMDF Chair Professor of Finance and Economics, Chair Professor of Finance and Economics
Abstract
Spot correlation is a key object for measuring high-frequency comovement across asset prices. We develop a new econometric framework for estimating spot correlation from high-frequency candlestick (OHLC) data. The proposed estimators combine open-to-close returns with wick-based asymmetry measures from paired candlesticks and correct the asymptotic bias in existing candlestick-based correlation statistics. Under a general continuous-time Itô semimartingale model, we establish consistency and large-k asymptotic theory, and we also develop fixed-k inference that remains reliable when the local estimation window contains only a finite number of candlesticks. A simple rule-of-thumb choice of the weighting parameter delivers substantial efficiency gains relative to conventional return-based estimators, while the fixed-k procedure affords more accurate coverage in empirically relevant sample sizes. An empirical application to intraday stock-bond comovement around FOMC announcements and tariff-related policy shocks illustrates the practical usefulness of the new methods.
Speaker
Prof. Qiyuan LI joined the HKU Business School in 2024 as an Assistant Professor. His research interests are in econometric theory. Currently, his research primarily focuses on financial econometrics with high-frequency data. He has published papers in Quantitative Economics, Journal of Econometrics, and Oxford Bulletin of Economics and Statistics, etc.
All are welcome!
