Max Bahar

This project explores how tropical cyclone strength evolves under varying climate scenarios in the Gulf of Mexico. Leveraging CMIP6 climate model data, we calculate the Maximum Potential Intensity (PI) using key variables like sea surface temperature and atmospheric conditions. The results highlight significant intensification trends, especially under severe climate scenarios, emphasizing the growing risk of stronger storms. The project includes a Python library, tc-potential, with tools for data fetching, analysis, and visualization, offering a streamlined workflow for understanding cyclone dynamics.

Check out the repository for more details and installation instructions!

Python

This project examines how demographics influenced voter turnout in Massachusetts during the 2020 election. Using interactive maps and a Random Forest model, it highlights key drivers like income, ethnicity, and language, revealing disparities in voter engagement. Explore the data, compare predictions, and discover actionable strategies to make voting more equitable.

Check out the interactive tool to dive deeper into the data!

Python

D3.js

Using earthquake data from the United States Geological Survey (USGS), I explore how Maptitude mapping software can be used to analyze the locations, frequency, and depth of earthquakes. As an example, I analyzed the relationship between the prevalence of earthquakes and the location of wastewater injection sites in Oklahoma, finding evidence of spatial correlation.

Check out the blog post to explore further insights!

Maptitude

I processed and integrated banking compliance datasets from the Community Reinvestment Act (CRA) and Home Mortgage Disclosure Act (HMDA), enabling financial institutions to gain actionable insights. The analysis-ready data at county and Census Tract levels supported visualizations of distressed areas, income distribution, and loan trends. Additionally, I developed workflows to streamline the integration of raw Loan Application Register (LAR) data for regulatory analysis.

Check out the product description to get the full details!

Python

Maptitude

To estimate the impact of the Albertsons and Kroger merger, I mapped Albertsons' and Kroger's locations (including their subsidiary brands), compared their geographic overlap, and estimated the population and demographics of those near the stores. I found that there were areas of significant overlap between the two companies (e.g. Los Angeles, Dallas, Washington DC) but that the companies' locations target different demographic segments.

Check out the blog post to get the full story!

Maptitude

For my undergraduate senior thesis, I sought to estimate the impact of parental migration on their children's future income. I used the Indonesian Family Life Survey (IFLS), a panel dataset examining the lives of 22,000 individuals over the span of 21 years and ran an instrumental variable regression using region-level migration rates to predict parental migration. Although the results of the study were ambiguous, with large standard errors affecting the coefficient of interest, I was able to show that a significant portion of parental migration in the IFLS was driven by non-labor reasons.

Check out the full paper to review the complete findings!

STATA