In June 2013, Calgary experienced extreme flooding, displacing 100,000 people and causing $1.7 billion in damages. This project aims to develop a predictive model to help Calgary and similar cities better prepare for such events. By analyzing predictors like elevation, land cover, slope, distance to water, and water flow accumulation, the model helps identify areas at risk of flooding beyond just riverine zones. The model's results aid in early action and planning, assisting decision-makers and the public in identifying areas needing evacuation, buyouts, or building code revisions. Using data from the 2013 Calgary flood, we also applied the model to predict flood risks in Denver, a city with similar geographic characteristics. This project was completed as part of the Land Use and Environmental Modeling course (CPLN 6750 / MUSA 6750) at the University of Pennsylvania. It demonstrates the application of spatial data and empirical models to support decision-making in urban planning and disaster preparedness.

This project was completed as part of the Public Policy Analytics course at UPENN. Emil City is considering a more proactive approach for targeting homeowners who qualify for a home repair tax credit program. This analysis aimed to train the best classifier to predict which homeowners are most likely to accept the credit and use the results to inform a cost/benefit analysis. By leveraging historical campaign data, we developed models to predict homeowners' likelihood of accepting the subsidy and conducted a comprehensive cost-benefit analysis. The goal was to optimize resource allocation and increase participation in the credit program.

This project was completed as part of the MUSA Public Policy Analytics course at UPENN, where students analyze real-world issues using data science tools. My partner, Nohman Akharti, and I collaborated to deliver comprehensive insights into the challenges and opportunities presented by Amsterdam’s tourist boom. By leveraging a variety of data sources, including Airbnb activity, housing market trends, and regulatory impacts, we developed models to predict the effects of different policy measures on local housing and tourism dynamics.

A Planning Support Tool for Affordable and Workforce Housing Site Identification in Watauga, North Carolina.