My research primarily focuses on the analysis of nonstationary time series data and its applications in economics and finance, including financial market forecasting, risk management, and systemic risk assessment. I am also dedicated to developing robust statistical methods for causal analysis.  Additionally, I explore applied econometrics across various domains in economics and finance, such as health policy evaluation, firm strategies, and natural resource management. My current research can be categorized into the following areas:Construction of Financial Market Systemic Risk Indicators: I am engaged in several projects aimed at constructing robust measures of systemic risk in financial markets. My research examines the roles of different markets within the financial system, including commodity, futures, and equity markets, and explores the relationships between them. By analyzing the interconnectedness of various asset types, I aim to develop risk indicators that can provide valuable insights for risk management and serve as early warning signals for policymakers and market participants.Investigating the Impact of Shocks and Key Factors on Market Systemic Risk: This line of research seeks to identify the impact of various shocks, such as Federal Reserve policy changes and the COVID-19 pandemic, as well as other key factors that contribute to the escalation of systemic risk in markets. I also explore potential transmission mechanisms to better understand the dynamics of systemic risk within financial markets.Developing Robust Instrumental Variable Estimation with Endogenous Instruments: In this area, I focus on improving instrumental variable (IV) estimation techniques, especially when instruments may be endogenous. One of my projects is closely related to recent advancements in the use of random forest techniques for causal inference. This work has broad applications in economics, particularly in scenarios where only one potentially endogenous instrument is available.Developing Robust Statistical Methods for Nonstationary Time Series Data: I am also working on innovative statistical methods for the estimation and inference of persistent, nonstationary time series data with nonlinear dependencies. Such data are common in finance and macroeconomic studies, especially when dealing with high-frequency data. My aim is to create tools that enhance the analysis and interpretation of these complex data sets.