Shanghai, China
•Excel, SAS, SQL, Python, PowerBI, VBA, Data Cleaning&Modeling, AB test, Decision Tree and Linear Regression •Proficiency in English and Mandarin; deep understanding in financial instruments and multinational e-commerce; well connected with retailers and wholesalers •Strong interest in philosophy, linguistics, computational logic, physics, badminton, hiking, photography and design
•Built brand image and connected with customers cross multiple industries such as printing, design, advertisement, banking, hospital and other commercial companies, and expanded business scope to over 50% of local wholesale and retail market while maintained annual outstanding account receivable ratios well below 5%. •Adopted a bottom-up analysis strategy that combines macro and micro perspectives by analyzing product reviews and volumes ordered by purchasing periods from various e-commerce platforms, focusing on understanding the consumption trends and purchasing power of markets at all levels in mainland China. •Assessed product selections, market shares and marketing campaigns of competitor companies using outsourced data. •Implemented AI, SQL, Python and Excel training sessions for employees in order to automate data processing and ultimately ensured legal and safe operations. •Assisted stock management through forecasting pulp future prices with linear regression models in order to smooth potential risks due to periodical price change in paper supply. •Cooperated with wholesalers through multiple e-commerce platforms: JD.com, ccgp-hebei.com, plap.mil.com, zkh.com, and responsible for sales, post-sale support, and deliveries at geographically granular levels.
•Assisted to complete 2018Q3 multinational general liability profitability study and 2019Q1-2021Q1 US segment general liability, auto liability and workers compensation profitability models to support product pricing strategy. •Reconciled and cleaned over 30 years of data from AIGRM through SQL and Python queries, and ultimately built a Power BI dashboard to capture multidimensional explanations of GMV change including loss trend analysis, business mix change, risk exposure, severity and frequency triangles and rate adequacy by year, state, industry, business and loss type. •Updated coal mining deep dives based on decision tree analysis including risk attributes of location, age, black lung disease, hypertension, COPD and kidney disease, and conducted decisive guidelines for both pricing and reserving. •Built web scraping tools with Python to reconcile and recategorize all insured companies by their core business in the past 10 years, and refreshed our business pricing models with more robust appreciation of industrial risks. •Proposed logical deficiencies in previous maturity risk premium calculation algorithm, and updated it with new linear regression models with Python (Pandas, Numpy, Scipy and Sklearn).
• Aggregated and analyzed data from Capital IQ by using SAS and VBA for all US public companies in the financial sector, including but not limited to merge & acquisition, security class action (SCA) claims, income statement, balance sheet and cash flow statement, daily traded volume and closing prices from past 10 years. • Implemented decision tree analysis with Python-sklearn for study of relationship between large financial claims and financial performance of insured companies, and successfully supported our strategic planning for the renewal accounts. • Gathered large claims by VBA from Financial Institution business segment including deductible, accrued loss and ultimate loss, and calculated corresponding IRIS (Insurance Regulatory Information Systems) ratios for each insured. • Conducted AB test and Z test on distributions of loss ratios among the insured companies by using SQL and Python to gain insight of possible correlation between loss ratio and abnormal IRIS ratio.