Front-end ScoreCard and Back-end XGBoost Credit Default Prediction Model
Data Mining and Application Course Project | May 2022 - Jun. 2022 | Prof. Jia Zhang | SWUFE
Collected Data and conducted data pre-processing and visualization (Python)
Developed credit default prediction models based on machine learning models with model evaluation by ROC, AUC, F1-scores by Python
Built the Credit Scorecard by logistic regression combined with XGBoost to predict the lender default likelihood, which is semi-interpretable and with enhanced generalization ability
Research on Central Bank Written Communication and Stock Market Volatility
Undergraduate Thesis | Jan. 2022 - May 2022 | Prof. Ziying Yang | SWUFE
Collected and pre-processed the text data of the minutes of the People's Bank of China's Monetary Policy Committee's regular meetings by Python
Used sentiment unit method and text analysis techniques to map the text to the time-series of tone scores
Constructed the EGARCH model to examine the impact of the tone scores of "meeting minutes" on the volatility of the Chineses stock market
Analysis of China's Foreign Exchange Market Daily Data
Data Analysis Project | Jan. 2021 - May 2021 | Prof. Jin-Ting Zhang | NUS
Attended weekly seminars covering data analytics and visualization topics
Completed data analysis project analyzing daily market data of China’s Foreign Exchange Market (R)