Data Enthusiast
Honors Data Science & Economics @ George Washington University
Rohan Weyerbrock Arora
Navigation
Technical Skills: RStudio, Python, SQL, Stata, Tableau, Power BI, Excel, MongoDB, Looker, dbt
Libraries/Packages: NumPy, Pandas, Matplotlib, Seaborn, Plotly, Statsmodel, Scikit-learn, dplyr, ggplot2, tidyr, shiny, astsa, RJAGS, MCMC
Education:
| BS Data Science & Economics | George Washington University Honors |
Work Experience:
Data Analytics Engineer Intern @ CarGurus (June 2024 - August 2024)
As an intern with the Data Engineering Team, I collaborated with Data Analysts, Software Engineers, and Data Scientists to build and maintain numerous data models using dbt, Snowflake, and Looker. I optimized SQL queries for datasets with +100 million rows, achieving high performance and cost reduction. Additionally, I designed and implemented scalable big data pipelines using Airflow, improving the efficiency of data flow and processing.
Data Science Year-Round Intern @ Federal Reserve Board (May 2023 - May 2024)
During my internship at the Federal Reserve Board with the BOPA team, I had the opportunity to hone my technical skills while contributing to impactful financial decision-making processes. I developed various machine learning models using Python and R, such as a predictive model for the division’s budget. I also created and maintained comprehensive Tableau and Power BI dashboards, providing managers/officers with clear insights that led to cost-saving opportunities. This experience was invaluable in solidifying my interest in data and demonstrating the real-world impact of analytical solutions.
Honors Research Assistant @ George Washington University (October 2022 - May 2023)
As a research assistant, I have experience constructing and analyzing large datasets to study complex topics. In this role, I created a panel dataset covering over 3,000 counties and 22 years to investigate the impact of unionization on voting access measures in the United States. I used quasi-experimental methods and regression analysis (using R and STATA) to study this relationship and conducted thorough research on the topic to validate and support my data analysis. This role interested me as it allowed me to develop my data skills, which I will use in my career.
AmeriCorps Member @ Mission: St. Louis, Beyond School (August 2020 - June 2021)
I served for an AmeriCorps Program called Beyond School Mission St. Louis. I helped a group of 13 sixth graders prepare for High School and beyond with social/emotional learning, enrichment activities, community service, and individualized academics. I was also the volunteer specialist of my school, helping recruit and train 40 volunteers to work with our students. I learned a lot while working for Beyond School, specifically skills of volunteer management, tutoring, and database management. The entire experience was really precious as it broadened my views and taught me important life skills, which I know I will use in the future.
Projects:
Personalized Book Recommendations Using Retrieval Augmented Generation (Python)
This project is a content-based book recommendation system that uses large language models to suggest titles based on user-inputted descriptions and emotional tone, rather than user history. Built with Python, LangChain, Hugging Face, and Gradio, it integrates zero-shot genre classification, sentiment analysis, and vector similarity search to deliver nuanced recommendations through an interactive web interface.
Read more about this project here
From Tweets to Foreign Trade: How Trump’s Social Media Shaped International Markets (Python)
This project investigates the impact of Donald Trump’s Twitter activity on short term foreign exchange market volatility during his presidency, employing high-frequency intraday data and utilizing the Anthropic Claude Sonnet 3.5 model for tweet classification. The study analyzes over 56,000 tweets, categorizing them into trade policy, international relations, economic policy, and security topics while examining their effects on major currency pairs, including the Turkish Lira, Canadian Dollar, and Mexican Peso. Through Ordinary Least Squares regression analysis and sentiment interaction terms, the research reveals significant correlations between Trump’s tweets and currency volatility, demonstrating how social media communications from world leaders can create distinct information shocks in global financial markets.
Read more about this project here
Shaping a Happier World: Insights & Regression Analysis on the 2017 World Happiness Report (R)
This project analyzed the 2017 World Happiness Report in R, examining the factors influencing happiness scores and variations across regions. The study utilized correlation matrices, box plots, regression analysis, and feature importance tests to identify that life expectancy, economic factors, and freedom were the most impactful variables on happiness scores, with regional variations observed in their correlations.
Read more about this project here
Multi-Linear Regression Exploring the Link between Food Insecurity and Asthma: Evidence from the 2018 National Health Interview Survey (STATA)
Through logistic regression modeling in Stata, I analyzed 2018 National Health Interview Survey data to quantify the relationship between food insecurity and asthma prevalence in the US. By constructing control variables for income, family size, and regional factors, I identified a statistically significant positive correlation between food access limitations and asthma risk. I verified my findings through hypothesis testing and robustness checks, addressing potential endogeneity concerns to validate the observed relationship. This econometric approach revealed important socioeconomic mediators of the food insecurity-asthma connection, contributing meaningful insights into health disparities while strengthening my capabilities in statistical inference and complex data analysis.
Read more about this project here
Additional Resources
Connect with me on LinkedIn