Daily prices may be noisy but volatility is less so. This project focuses on forecasting how much NVDA is likely to move tomorrow using an EGARCH baseline model as well as an XGBoost hybrid model. It also considers two strategies and how they perform using forecasted volatility targets.

This project uses bagging and random forests to predict “high matchability” from OkCupid-style profile data, and compares performance against a single decision tree baseline. Alongside the predictions, it also highlights the strongest drivers of matchability (values, lifestyle, life-stage signals).

This project explores a PCA-driven deep dive into how people perceive fictional characters, using hundreds of crowd-rated personality traits and compressing them into a few clear “personality dimensions.”

Using Spotify audio features, we clustered years (1921–2020) to see whether music history naturally breaks into distinct “eras” based on how songs sound. Hierarchical clustering revealed three clear periods (Early, Mid, Modern) with noticeably different energy, acousticness, loudness, danceability, and popularity profiles.