About me
I am a graduate student in the Statistical Data Science program at San Francisco State University
Areas of interest:
Social Welfare
Clean Energy
National Defense
This study, which serves as my Master's thesis, addresses the lack of standardized reporting for the 20 million annual traffic stops in the United States by modeling stop demographics at the county level using census and geospatial data. By applying statistical distance measures and a custom geospatial smoothing procedure, the research quantifies and visualizes disparities between local populations and those being stopped. The findings reveal that these demographic disparities vary significantly by race, sex, and police department type rather than scaling uniformly with population size. Ultimately, this project provides a data-driven framework to better contextualize traffic enforcement trends within a broader national landscape.
To provide a little perspective on just how much money a billion dollars is, this tool provides a series of visualizations of the scale of the wealth of the world's richest people were it to be held in physical currency, specifically, in some denomination of dollar bills. Also, just for fun, you can compare the world's current 25 richest people with some of the richest individuals in history, excluding rulers and others who derived their wealth from their positions in government, and the richest fictional characters ever. Last updated in 2023.
This tool finds the titles of movies or names of actors and directors in the transcripts of The Big Picture podcast that are available on Spotify. Movie titles are matched with Wikipedia's list of films and actors and directors are matched with Wikipedia's lists of actors and directors who have been nominated for an Academy Award.
Crossword puzzles built entirely with clues and answers that are movie-related. Crossword construction is automated using python with a unique clue/solution dictionary augmented by interaction with a local language model.
Compare the null and alternative hypotheses when testing multiple means with a graphic representation of how the distributions overlap according to the number of treatments, number of samples per treatment, means of the treatments, and the overall variance. It also displays the areas corresponding to Type I and II errors and calculates the Beta and Power of the experiment.