Python Data

XINRAN (VIVIAN) TIAN

** ***** ****, **, ***** 631-***-**** *******@*******.***

https://www.linkedin.com/in/xinran-tian/

EDUCATION

FORDHAM UNIVERSITY, GABELLI SCHOOL OF BUSINESS New York, NY MS, Quantitative Finance, GPA 3.82 Aug. 2019-Expected May. 2021

• Coursework: Fixed Income, Data Infrastructure (SQL), Data Visualization (Tableau), Web Analytics, Machine Learning, Natural Language Processing

STONY BROOK UNIVERSITY Stony Brook, NY

BS, Applied Mathematics and Statistics, GPA 3.7 Aug. 2015-May. 2019

• Coursework: Corporate Finance, Linear Algebra, Stochastic Calculus, Linear Algebra, Econometrics TECHNICAL SKILLS

• Scripting Languages: Python (NumPy, Pandas, SciKit-Learn, MatplotLib, Seaborn, BeautifulSoup, NLTK), SQL

(MySQL, PostgreSQL, AWS Redshift), R (randomForest, xgboost, ggplot2)

• Statistical/ML Techniques: A/B Testing, ML (SVM, Decision Tree, Random Forest, XGBoost)

• Visualization Tools: MS Office, MS Excel/Pivot tables, Tableau

• Data Management: AWS (EC2), Apache Spark

WORK EXPERIENCE

RECURSION CO. New York, NY

Data Analysis Intern (SQL, Python, Tableau, Excel) Jun. 2020-Present

• Handled ad-hoc requests from cross-functional teams, including using SQL to extract over 5,000 raw mortgage data points; built Tableau dashboards to help R&D department track data performance

• Customized data quality check models in Python and lessened R&D department workload by 50%

• Developed Random Forest Model over 306K data by using Python to forecast mortgage delinquency rate (DQ rate), and analyzed influences of features on DQ rate; illustrated and visualized results in company’s weekly blog

• Adhered to principals of data quality to redesigned automated web scraping program to scrap more than 3M data points; boosted teamwork efficiency and reduced manual work by 70% FOUNDER SECURITIES CO., LTD Beijing, China

Quantitative Analysis Intern (SQL, Python, Excel) Jun. 2018-Aug. 2018

• Developed risk evaluation databases in PostgreSQL, yielded risk data tracking effectively up to 80%

• Supported Developer implement an end-to-end pipeline for ETL; reconstructed daily risk report by improving model using Python; simplified data-reading process in databases

• Accelerated debug process by regular data extraction; presented error analysis report to cross-functional teams

• Built and tested trading strategy by implementing Risk Parity Method on adjusting factor weights of a multi- factor model; increased Sharpe Ratio by 1.8% using the updated strategy ACADEMIC PROJECTS, FORDHAM UNIVERSITY

LOAN DEFAULT PREDICTION – Kaggle Dataset (Python) May. 2020-July. 2020

• Performed Exploratory Data Analysis (EDA) on 8,000,000 observations with 47 features (includes 15 anonymous) and constructed Features Engineering for both categorical and numerical features

• Developed and tuned XGBoost and LightGBM models separately to detect personal loan default; merged three models by using arithmetic mean algorithm; enhanced AUC score from 0.66 to 0.7 MACHINE LEARNING FOR PORTFOLIO TRADING DESIGN (R) Oct. 2020-Dec. 2020

• Applied Lasso regression model as quantitative strategies to construct a 500 stocks equity portfolio; used Random Forest models to sift stocks and selected profitable stocks as first week’s portfolio

• Generated portfolio rebalancing signals for every week with above Lasso regression to outperform the S&P 500 index; evaluated the performance of portfolio returns to actual stock price movements

Contact this candidate