Real-time Monitoring in Panel Quantile Models with Application to Global Inflation
Holder： Yiren Wang（Hunan University）
Time：2026-05-21 15:10-17:00
Location：Room 217, Guanghua Building 2
Abstract:
This paper develops a real-time monitoring econometric framework for detecting structural changes in global inflation heterogeneity. The framework is designed to detect changes in the inflation-macroeconomic relationship across quantiles in a timely manner as new data become available. We employ a convolution-type smoothed panel quantile regression model and construct a gradient-based sequential detector with an associated boundary function. The proposed procedure preserves convexity and achieves -consistent estimation through bias correction. We establish asymptotic size control under the null hypothesis of parameter stability and show the non-trivial power under local alternatives. Simulation verifies the theoretical properties of the proposed method. Applying the method to a novel monthly panel of 43 countries, we show that real-time monitoring detects changes in global inflation dynamics around the Global Financial Crisis and COVID-19 pandemic: the financial crisis episode shows detectable but heterogeneous breaks, whereas pandemic episode yields concentrated real time detections and stronger sensitivity to real economic activity at lower and middle quantiles after the break.
About the Speaker:
王奕人，湖南大学金融与统计学院助理教授。研究方向为面板数据、高维计量经济学等。研究成果发表于Journal of Econometrics，主持国家及省部级自然科学青年基金共两项。
Your participation is warmly welcomed!
欢迎扫码关注北大统计科学中心公众号，了解更多讲座信息!
