August 2025
Machine Learning in Financial Risk
Data without Borders: Building AI Competence across Borders · Romania (opens in a new tab)
Summer course
Classroom
Teaching experience
Feb–Sep 2021
DEDA — Digital Economy & Decision Analytics
Humboldt-Universität zu Berlin
Sep–Dec 2019
Probability Theory
Sun Yat-sen University
Sep–Dec 2019
Financial Markets and Financial Institutions
Sun Yat-sen University
Open learning
Quantinar courses & courselets
Q / 01
Machine Learning in Financial Risk
Introduces VaR, Expected Shortfall, quantile regression and Lasso, then connects them through Financial Risk Meter applications.
Open course (opens in a new tab) Q / 02FRM@China
Develops FRM@China to detect systemic financial risk, tail-event dependence and risk drivers across major Chinese financial institutions.
Open course (opens in a new tab) Q / 03FRM@Crypto
Applies the Financial Risk Meter to cryptocurrencies and links the FRM index to pricing-kernel volatility and forward-looking risk.
Open course (opens in a new tab) Q / 04Measuring Carbon Risk Dynamics
Develops CARDI, a Quantile-LASSO network indicator for carbon-risk transmission, volatility shifts and low-carbon premia.
Open course (opens in a new tab) Q / 05Market Responses to ETH Development Milestones
Studies how Ethereum upgrades affect returns, volatility and spillovers across the broader cryptocurrency market.
Open course (opens in a new tab) Q / 06Kalman Filter
Introduces Kalman filtering as a way to infer unobservable states from observable data, with applications to financial time-series prediction.
Open course (opens in a new tab) Q / 07Applied Time Series Analysis Solutions
Collects solution materials across time-series concepts, ARMA/ARIMA, nonstationarity, financial time series, state-space models and modern ML methods.
Open course (opens in a new tab) Q / 08Pricing-kernel puzzle
Explains the empirical pricing-kernel puzzle by comparing risk-neutral and physical densities, preference implications and volatility-driven evidence.
Open course (opens in a new tab) Q / 09Quantile Regression
Introduces quantile regression for conditional distribution and tail analysis, covering asymmetric loss functions, estimation and empirical applications.
Open course (opens in a new tab) Q / 10Expectation Maximisation (EM) algorithm
Explains the EM algorithm for maximum-likelihood estimation in probabilistic models with missing data or latent variables.
Open course (opens in a new tab) Q / 11Bahadur, Edgeworth, Cornish-Fisher expansions
Covers Bahadur, Edgeworth and Cornish-Fisher expansions for improving normal approximations and quantile-based risk estimation.
Open course (opens in a new tab) Q / 12Bootstrap and Jackknife
Introduces bootstrap and jackknife resampling methods for estimating bias, standard errors and uncertainty in statistical estimates.
Open course (opens in a new tab) Q / 13How Much Power was Saved in ETH Merge?
Uses the Ethereum Merge as a natural experiment to measure cryptocurrency energy consumption with a hybrid framework.
Open course (opens in a new tab)