Jerry Hong
Welcome to my Portfolio!
I am Jerry Hong, and my passion is to leverage various technologies to uncover hidden insights and drive decision-making. In the growing digital world with enormous amounts of data, my passion is continue building my technical and critical thinking skills to help businesses and organizations make decisions through the data collected. One of my growing interests is pursuing economic research that can influence decisions in policy work. In addition, I look to continue building my toolkit of various software that handle working with large amounts of data, from cleaning to wrangling to visualizing. Check out some of my favorites projects below showcasing my skills within data science.
Education
-
M.S., Applied Economics Emory University (May 2025) -
B.A., Economics Emory University (May 2024) -
B.S., Data Science Emory University (May 2024)
Certifications
- Tableau Desktop Specialist
- AWS Certified Cloud Practitioner
DataCamp Portfolio
Check out my portfolio courtesy of DataCamp, which is a great site for building my data science skills!
Projects
Replicating a Study on Price Discrimination through Negotiation
This project is an attempt to replicate the findings that are carried out in the study of price discrimination by negotiation. The premise of this study is to study price discrimination in a market with price posting and negotiation. The motivation stems from the concerns that low-income consumers do poorly in markets with privately negotiated prices. As such, the authors carried out a call center with actors engaging with the electricity and utility service vendors throughout Victoria, Australia. With different scripts characterizing people of various income status, the goal of the study is seeing if there are differences in the outcomes of price invoices between low and high-income individuals. We will attempt to replicate the findings from their research through the provided replication package.
Instrumental Variable Simulations
This project demonstrates my understanding of the use of instrumental variables in causal research studies. Using the given context, I explore the incorporation of instrumental variables that may influence the treatment effect which in turn can affect the outcome variable. Using a data generating process, I simulate the process as well as running different regressions that may determine the strength of these instrumental variables.
Power Simulations
This project showcases running a power simulation and understanding the relationships between power, sample size, and minumum detectable effect in causal research studies. The following is the abstract: In statistical inference and hypothesis testing, understanding the dynamics of sample size, power and effect size are important when conducting such studies. Each of these attributes directly influence one another through both direct and inverse means. What follows is a series of power simulations showcasing the relationships among these factors through both a synthetic data generation and a household survey data set. These simulations can allow us to find one of the unknown factors when the other two are known. Mainly, finding the minimum detectable effect (MDE) and the sample are most common to find through these simulations.
Developing a Research Design
This project demonstrates a surface-level process of analyzing and developing research designs. Part I focuses on formulating research designs by: identifying the dependent and independent variables, observation and periodical variables, and null and alternative hypotheses. Part II focuses on basic data and regression analysis of health and labor economics data. We will to the best of our ability replicate regression figures from each of these studies.
Help a Non-profit Organization Expand their Community Outreach
Friends of Disabled Adults and Children, or FODAC is a statewide organization based in Georgia where their primary mission is making expensive medical equipment more accessible for those in need. Their goal is to acquire grant funding to expand their business outside the Atlanta-metropolitan area. Our team was tasked to provide data insights highlighting their contributions to the community as well as exploring target communities for outreach. Above is a presentation sharing our findings as well as dynamic dashboard for FODAC to update with new data.
Using Machine Learning Techniques to Predict Airbnb Prices in New York City
This project showcases my understanding in machine learning by utilizing the techniques that best fit the scope of our research. The goal is to explore which models can best predict prices for Airbnbs for consumers to make better informed purchasing decisions and maximizing the value of the living accommodations. We also want to assess whether the predicted prices relative to the actual prices can be attributed as a good deal or not. My part of the project is by testing out OLS linear regression, decision tree, and random forest models, where the third option ended up being the best performant. In addition, I also assessed additional factors and further questions to expand in our study.
Exploring Major Contributors in Housing Valution using Regression Models
Leveraged Python and analytical skills to analyze IPUMS housing data, a rich dataset encompassing various socioeconomic factors. Utilizing regression and statistical modeling techniques, I developed predictive models to forecast housing valuation trends, exploring major economic contributors that most influence housing values. Created a comprehensive Jupyter Markdown report, highlighting key factors influencing housing valuation and demonstrating a nuanced understanding of market trends within the housing sector.
Finding the Best-Value Accommodations in Venice, Austria using Data Analysis
Analyzed consumer data of various living accommodations throughout Venice, Austria. Used R to clean, create new variables, and visualize data highlighting factors that can affect the pricing of accommodations. This can provide additional insights for prospective travelers in search for the best deals to maximize their vacation experince. Created an RMarkdown report to showcase my methodologies and findings.
An Evaluation of Pennsylvania’s Re-employment Incentive Experiment
The purpose of this experiment is to evaluate the effective of a re-employment bonus program through various incentive initiatives. The goal of this evaluation is to explore demographic factors that most impact the program across three categories: race, age, and gender. I conduct a series of regression tests to calculate the average treatment effect of these demographics as well as the possibility of interactions between variables. My takeaways are that the program produces a positive impact on the job outlook of unemployed workers. While the variables have contributed to the study, the interactions with each other remain to be seen, implying uniformity among the test groups. Used R and created an RMarkdown report throughout this research.
Using Python, TabPy, and VSCode to Leverage your Data Science Workflow
Created a Markdown guide for beginners installing essential software like Python, Tableau, and VSCode. The guide also features an example of integrating Python with Tableau to run scripts within Tableau. This can help facilitate aspiring data scientists and analysts to get started in their data science journey.