Quantitative researcher

Yue Ma

(***) *** – **** adlmmu@r.postjobfree.com

Newark, New Jersey

EDUCATION

Rutgers Business School, Newark, NJ Aug 2018 – May 2020 Master of Quantitative Finance

Core Courses: Optimization Models, Stochastic Calculus in Finance, Time Series Econometrics, Financial Modeling of Options, Object- oriented Programming, ETF & Indexing, Numerical Analysis, Panel Data University of Cincinnati, Cincinnati, OH Aug 2014 – May 2018 Dual degree: Finance, B.B.A. & Mathematics, B.S.

Honors: Global Scholarship, Sarah Blank Greenholz Scholarship, Math S.T.E.M. Fellowship, Dean’s List Work Experience

Fixed Income Quantitative Finance Researcher Oct 2020 – Present Basis Point Global Solutions, LLC., Newark, NJ

● Developed optimal asset allocation across fixed income products by using mean-variance optimization in Python with Scipy

● Improved the optimization results by applying exponentially weighted model on the covariance matrix of assets, adding futures into the portfolio and incorporating the credit migration approach

● Constructed a core plus index from fixed income products with target tracking errors, with in sample results that outperformed the benchmark Bloomberg Barclays US Aggregate Statistic Index (LBUS) for 15 years

● Predicted fixed income index total returns from macroeconomic data with machine learning models such as KNN, Random Forest regression and SVM as well as deep learning models such as LSTM in Python with Scikit-Learn, Keras &TensorFlow Data Analyst Intern Oct 2020 – Jan 2021

Concord Advice, LLC., Newark, NJ

● Reduced dimentionality by extracting strong and weak features from over 17,000 raw bank transactions data

● Trained Random Forest and Gradient Boosting models to construct classifers for bank transactions with AUC of 0.97, precision of 0.967 and recall of 0.935

● Developed a probability pricing model that maximize daily profits for personal loan offering; Used information ratio and Naïve Bayesian classification to predict the probabilities of loan conversion (accuracy of 0.4) and defaults of customers (accuracy of 0.83)

Professional & Academic Projects

Stock Sentiment Analysis on Financial News Headlines (NLP, Python) Aug 2020 – Sept 2020

● Scraped news headlines for given stocks with Python packages BeautifulSoup & Urllib from Finviz.com and performed the VADER sentiment analysis with NLTK

● Visualized results with Plotly and produced charts such as sentiment scores for a single day trading & a single stock, positive versus negative sentiment, and stock price movements versus stock news sentiment scores Creation of Index in the Pharmaceuticals and Biotechnology (Excel, Python) Jan 2020 – May 2020

● Defined a universe of 85 stocks with FDA approved patents in chronic diseases with a team of 4

● Applied 4 different weighting schemes in index creation and backtested the strategies for the past 20 years, with the final index returns beating NASDAQ Biotechnology Total Return Index (XNBI) and Dow Jones U.S. Select Pharmaceuticals Total Return Index (DJSPHMT) for the past 5 years

● Wrote an index rule book and presented the project to 7 industrial experts in ETF and index funds Model Fitting to Data on Default of Credit Card Clients (Python) Oct 2019 – Dec 2019

● Applied feature creation & standardization techniques to preprocess data and evaluated new features

● Trained KNN, logistic regression, decision tree and neural network models on data in python with Scikit-Learn. Compared precision, recall, accuracy, and mean square error across models and found that a post-pruning decision tree had the best fitting with the smallest mean square error of 0.183 and the highest predictive accuracy of 0.817 Detection of Cryptocurrencies Bubbles with Econometrics Models (R, Stata) Feb 2019 – Apr 2019

● Built various econometrics models, such as the OLS model, the IV model, the random effect model and the fixed effect model, to analyze the correlation of economic factors and prices of 4 cryptocurrencies

● Tested models assumptions with t-test, F-test, Ramsey RESET test, White test, Durbin Watson test and etc, and found obvious evidence of bubbles in Bitcoin and Monero

CERTIFICATES & SKILLS

Certificates: CFA Level II Candidate (7309676), Exam Date: Sep. 2021 Quantitative Methods: Equity Trading Strategies, Model Calibration, Stress Testing, Data Analysis, Statistical Modeling Machine Learning, Deep Learning, NLP, Sentiment Analysis, Tensorflow, Keras, PyTorch, Pandas Computer Skills: Python, R, C++, MATLAB, Bloomberg, SQL, Excel VBA, LaTex

Contact this candidate