Sign in

R, Python, SQL, Data Modeling, Statistics

New York City, NY
February 26, 2020

Contact this candidate


Ting-Hsuan Tina Yang

Queens, NY ***** 214-***-**** Summary

Quantitative and analytical professionals with demonstrated experience in applying data analytics to solve financial problems and make predictions to better decision-making. Creative self-starter with established capabilities that solve complex problems. Passionate about employing data mining to develop applications of problem-solving, trading and risk management. Core Competencies

• Analysis: Textual Analysis, Natural Language Processing, Quantitative Analysis, Financial Modeling, Statistical Modeling, Machine Learning, Regression Analysis, Decision Trees, Classification System, Monte Carlo Simulation

• Technical: R(ggplot2, caret, rpart, tm, tidytext, coreNLP) (5 years), Python(pandas, numpy, scikit-learn, matplotlib, nltk, bs4) (4 years), C++(1 year), Matlab(2 years), SQL(1 year), Tableau(1 year), Microsoft Office, Bloomberg(basics)

• Certifications: CFA Level I

• Language: Chinese(native), English(fluent)


Columbia University, MS in Applied Analytics; GPA 3.75 New York, USA

• Relevant Coursework: Applied Analytics Frameworks and Methods, Strategy and Analytics, Dec. 2019 Machine Learning: Concepts & Applications, Python for Data Analysis, SQL and Relational Databases National Tsing Hua University, MS in Quantitative Finance; GPA 3.86 Hsinchu, Taiwan

• Relevant Coursework: Derivatives Pricing, Financial Risk Management, Numerical Methods in Finance, Jun. 2017 Data Mining Research & Practice, and Financial Texture Analytics

• Achievement: Received scholarship from Concord Futures and Taiwan Futures Exchange in an options trading contest. National Chung Cheng University, BA in Economics; GPA 3.84 Chiayi, Taiwan

• Relevant Coursework: Econometrics, Mathematical Statistics, Mathematical Economics, and Game Theory Jun. 2015

• Achievement: Won the Academic Excellence Award twice. Professional Experience

BitMart Exchange New York, USA

Research Intern Apr. 2019 - Sep. 2019

• Researched Bitcoin’s market growth, price cycle, and statistical characteristics to understand the investment side of Bitcoin.

• Established data analysis on the quality of initial coin offering (ICO) with web crawling to select qualified cryptocurrency.

• Devised a strategy of arbitrage opportunities for Bitcoin between cryptocurrency exchanges with ccxt module in Python. National Tsing Hua University Hsinchu, Taiwan

Teaching Assistant- Master Program of Finance and Banking Sep. 2016 - Jan. 2017

• Presented the topics of text mining with investment trading strategy and GUI based on R programming language.

• Guided students in solving problems in R programming language in class and through group chat.

• Edited Mainland Chinese financial word list to make it applicable to the text mining in Traditional Chinese. Graduate Projects

Capstone Project: Real Estate Property Foreclosure Predictive Model (Python, AWS) May. 2019 - Aug. 2019

• Achieved the accuracy rate of 91% when building the logistic regression model to forecast foreclosure with property data.

• Clarified what household characteristics are leading to foreclosure by creating factor tables for 21 screened predictors.

• Launched feature selection with domain knowledge, correlation matrix, univariate and forward selection.

• Handled the imbalanced data by applying oversample method and evaluating the model performance with precision. Text Mining on Investment Trading Strategy: An LDA Approach to Distinguish Topics (R) Mar. 2016 - Jun. 2017

• Inspected and deleted sentiment words which are not related to business contextually to properly measure the sentiment.

• Used Latent Dirichlet Allocation (LDA) to filter the non-business topics in the minutes of Federal Open Market Committee.

• Predicted the opportune moment to invest in S&P 500 based on the business-related sentiment indicator.

• Improved the annualized return rate by 0.8% after including LDA into the trading strategy with 12 transactions in 8 years. Death Prediction on Characters in “Game of Thrones” (R, Python, Weka, Spark, and Hadoop) Sep. 2016 - Jan. 2017

• Built a decision tree model with entropy to predict the death of characters based on their nobility status and death record.

• Promoted the accuracy rate to 68% by remaining the same ratio of dead and alive in both train and test dataset.

• Applied Spark (RDD, MLlib) and Hadoop to write and expedite the program process. Leadership Activities

• Leader of Art Design, National Chung Cheng University Student Association Sep. 2013 - Jun. 2014

Contact this candidate