A recent graduate with a master's degree in Statistics, I enjoy creating nuanced insight from data. My goal is to inform and improve institutional decision-making. Trained in multivariate data analysis, predictive modeling, computational methods for optimization, and machine learning techniques. Experience in data cleaning, visualization, and analysis.
Outside of my career: I enjoy learning about different cultures, cooking, dog-walking, and swimming.
The survey data, ICPSR 36536, is maintained by the Inter-university Consortium for Political and Social Research from the University of Michigan. The analysis I conducted for my Machine Learning course was motivated by the desire to understand who is at greater risk of using and abusing specific drugs. The analysis identifies clustering of the data and creates models for prediction, specifically whether someone is a user of a drug in the past decade. See more details here.
This is a selection I wrote for a group project on Hamiltonian Monte Carlo (HMC). Briefly, the HMC uses physical system dynamics to predict future states in the Markov Chain. The trajectory found by these Hamiltonian dynamics is evaluated by the “leapfrog” symplectic integrator. The proposals are then selected using standard Metropolis procedure. The acceptance rate is typically very high (>70%) compared to other methods of deriving samples like standard Metropolis or Gibbs. See more details here.
These are books I highly recommend regarding statistics and analysis:
Downey, Allen B. Probably Overthinking It: How to Use Data to Answer Questions, Avoid Statistical Traps, and Make Better Decisions. Chicago, USA: University of Chicago Press, 2024.
