Research

The Childcare Effect on Homelessness

Explorers the relationship between the Universal Pre-K rollout in New York City with 4K in 2014 and 3K in 2019 and the impact on levels of sheltered homeless families.

Used a difference-in-difference model to test if the expansion of early education programs in New York City reduces the population of homeless families with children in the affected group. Used publicly available data from various sources, including the NYC Mayor’s Office of Management and Budget, the Department of Homeless Services, and the Department of Education. Data sets were aggregated and cleaned in Excel and analyzed in Stata. The paper’s main limitation is the aggregate macro nature of the publicly available data. I also completed Human Subjects training and went through the IRB process to interview a small set of homeless individuals for a qualitative section of the paper. Advised by Professor Dan O’Flaherty.

This paper is still in progress as I work with the City of New York to obtain de-identified individualized homeless intake data from 2010 to 2022. This will refine my regression equations and results to help determine if universal childhood education policy impacts incidence of homeless families. The December 2, 2022 edition was the turn-in edition for my master’s thesis. Subsequent editions are updated versions of the paper.

Monopoly Money

Explores relationship between market concentration and equity returns in the U.S. stock market between 2001 and 2020. Aggregated data from COMPUSTAT and CRSP databases, creating a Herfindahl-Hirschman Index from North American Industry Classification System (NAICS) codes.

As a primary author and coder for this financial economics paper, merged two Wharton Research Data Services (WRDS) databases – COMPUSTAT and Center for Research in Security Prices (CRSP) listing 13,000 publicly traded U.S. firms from 1987 to 2020, which we organized by their NAICS code. Based on a company’s industry (NAICS code) market share, we built a Herfindahl-Hirschman Index (HHI) to rank a firm’s industry concentration for a particular year. We then use the Fama-French three-factor model and Carhart four-factor model to test if there was a significant difference in stock market returns between the top 20% of HHIs and the bottom 20% of HHIs. All cleaning and data analysis was completed using SAS and SQL. We were able to duplicate much of the previous research showing a correlation between low HHIs and higher returns prior to the year 2000. In the early 2000s the dynamic between market concentration and portfolio returns shifted dramatically, as companies with higher HHIs began exhibiting higher returns than those with low HHIs. Advised by Professor Steven Ho.

Research Assistantships

  • (May 2022 – Present) Content developer of R Shiny web-interactive modules for Dr. Rajiv Sethi. Our goal is to bring interactivity to CORE’s The Economy textbook, to make it more accessible to students. I and another RA have programmed four modules thus far and plan to complete the remainder of the modules in 2023. An example module is https://interactive-economics.shinyapps.io/MeasuringInequality/. I can provide further modules and code upon request.
  • (January 2022 – May 2022) Data cleaner for BooKang Seol’s Ph.D. dissertation, studying poverty alleviation impacts of 1970s community-driven development projects in South Korea. All work was done in Microsoft Excel, organizing and cleaning data for regressions which were run in Stata.