Patty Bi

Evaluated 900 risk cases, utilizing a risk matrix to target high-risk scenarios, while engaging with senior management to enhance risk identification methods. Aligned strategic initiatives with corporate goals, and ensured accuracy in large-scale operations by employing MySQL for efficient data handling. Worked collaboratively with 10 employees.

Processed large volumes of customer data using Python and performed Exploratory Data Analysis (EDA) to uncover key insights. Designed data visualizations in Tableau to support data-driven decision-making. Provided risk alert reports to the business team with actionable recommendations. Managed and guided a team of five employees, ensuring timely delivery of high-quality outputs

Conducted due diligence on the target company, utilizing both quantitative and qualitative factors to perform fundamental credit analysis. Reviewed the target company's balance sheet, cash flow statement, and income statement, and conducted an in-depth credit risk evaluation by analyzing key financial metrics. Collaborated with a seven-member team to enhance cross-functional communication through regular meetings, bridging the gap between technical and business perspectives

Collaborated with a team of 10 employees to conduct in-depth data analysis on auto loans and factoring, identifying patterns and trends to minimize high-risk loans. Worked closely with cross-departmental teams to verify loan amounts, ensuringaccuracy and compliance with financial regulations. Leveraged analytical insights to support risk management strategies and improve overall operational efficiency

Optimized workflows and improved the efficiency of consulting services through a shared platform and cross-departmental collaboration, streamlining operations. Implemented an electronic document management system to deliver advanced and comprehensive client services

## 2023-05 - 2023-06 Forecasting the inflation rate in the United States • Utilized ARIMA time series method to forecast U.S. inflation rate, selecting optimal (p, d, q) combination using information criteria, and conducted comparative analysis with triple smoothing model. ## 2023-10-2023-12 Forecasting the Heart-disease • Incorporating the AUC model evaluation method, we found random forest algorithm, achieving a 98% accuracy rate in predicting heart disease likelihood.