Machine Learning Approaches to Predict Crime Rate in the City of Los Angeles
Published:
Poster presentation exploring crime rate prediction in Los Angeles using multiple machine learning algorithms.
Portfolio
Predicting Loan Approval with LASSO Logistic Regression
Published:
End-to-end pipeline for loan-approval prediction: EDA, interaction diagnostics, LASSO feature selection with 5-fold CV, thresholding to control FPR.
Forecasting Used Cars & Trucks CPI: Regression vs. ARIMA
Published:
Comparative analysis of OLS, WLS, and ARIMA models to forecast U.S. Used Vehicle CPI. The study identifies ARIMA(0,1,2) as the optimal model for short-term prediction, outperforming macroeconomic regression approaches.
Bayesian Hierarchical Modeling of Serious Crime Rates (LAPD 2024–2025)
Published:
Hierarchical Beta-Binomial model implemented in JAGS to estimate serious crime probabilities across 21 LAPD areas, using Gibbs sampling to stabilize estimates for high-variance districts.