MINJIE DONG

848-***-****

** ***** ** *, ****** City, NJ, 07310 ac76z3@r.postjobfree.com

EDUCATION

Rutgers University, New Brunswick, NJ, USA

Master of Science, Financial Statistics & Risk Management, GPA 3.9/4.0 2016 – December, 2018 Tongji University, Shanghai, China (dual-degree program with EBS) Bachelor of Science, Accounting, GPA 4.13/5.00 2016 EBS Universität für Wirtschaft und Recht, Oestrich-Winkel, Germany Bachelor of Science, General Management, specializing in Finance, GPA 0.87/1.00 2015 PROFESSIONAL EXPERIENCE

Nomura Securities International, Inc. Greater New York City Area, New York Research Summer Analyst June 2018 - August 2018

§ Investigated the trend of US government deficit and developed a projection model using linear regression. The residual analysis revealed that tax-cut based fiscal stimulus resulted in larger-than-expected deficit to GDP ratios.

§ Examined long-term relationships among US treasury securities with different time periods using cointegration.

§ Automated a daily monitor of treasury securities to find cointegrated pairs and detect any short-term divergence in Python.

§ Conducted back-testing using Python to facilitate the successful implementation of pair trading strategy.

§ Built a daily sentiment index for online Fed news in Python utilizing the automated process such as web scraping and sentiment analysis with natural language processing methods to investigate its long-term relationship with US treasury securities. Pinz Capital Management, LP Greater New York City Area, New York Data Scientist Intern February 2018 – May 2018

§ Developed a natural language processing algorithm of stock collection for a thematic ETF in Python. The automated process included web scraping, name entity recognition, and sentiment analysis.

§ Applied logistic regression to train a text sentiment classification model in Python.

§ Used k-means clustering in article keyword extraction to build and maintain a news database in MySQL.

§ Developed a portfolio back-testing algorithm to adjust weights in rebalancing days and calculate Sharpe Ratio using Python.

§ Conducted an optimization model to meet user’s requirements for portfolio’s stock weights with Graphic User Interface in Python. Imagine Software, Inc. Greater New York City Area, New York Data Analyst Intern February 2017 – August 2017

§ Conducted standardized model in Basel’s Fundamental Review of the Trading Book for banks to manage market risk.

§ Modified and verified bond’s features in local database using Java-script API.

§ Used Java-script API to meet clients’ specific requirements, such as calculation of change on swaption to interest rate. RESEARCH PROJECTS

Explore Machine Learning methods for Credit Default Prediction 2017

§ Implemented traditional classification methods (e.g. logistic regression, linear discriminant analysis, k-nearest neighborhood, support vector machines, and random forest) to predict credit default for a dataset in Taiwan from April 2005 to September 2005, using R and Python.

§ Developed a stacking model with new features from the previous methods using logistic regression.

§ Compared with the other models by misclassification error, sensitivity and computation time, the stacking model performed best. Time Series Analysis for S&P 500 Index 2017

§ Conducted and compared time series models using R to analyze tendency of S&P 500 Index.

§ Used ARMA model to remove minor linearity in log return of the index. Found that ARMA (1,1) and MA (5) are preferred models.

§ Utilized GARCH model to fit non-linearity in the series and adjusted the candidate models. Found that ARMA (1,1) – GARCH (1,1) under skewed t-distribution performed best in rolling forecast. Identifying Non-Systematic Factors That Influence Stock Returns 2016

§ Developed regression models using R to identify factors that impact stock returns for companies in the S&P 500 Index.

§ The preferred model included systematic, categorical (e.g. Industry and Seasonal factors) and technical factors (e.g. MACD, Momentum and ROC); the technical factors were included as the difference between the current and previous period values.

§ Examined the performance of the preferred model by predicting stock returns for the next period; as measured by Sharpe Ratio, the preferred model outperformed the S&P 500 over the same period. TECHNICAL AND OTHER SKILLS

Core Domain Expertise: Machine Learning, Quantitative Analysis, Financial Data Analysis, Natural Language Processing Computing and Programming: R, Python, Java-script, MySQL, SPSS, Eviews, Microsoft Office Suite Communication and Teamwork: Fluent in English, Mandarin, basic in German, self-directed, enjoy team work HONORS AND ACTIVITIES

Affiliate Membership: GARP, IAQF, BMC, CFA Level 2 Candidate 2017

Contact this candidate