As a bright-eyed undergrad with no idea what "programming" meant, nor "statistics" or "data analysis", I was understandably intimidated when I started the "Scientific Data Analysis with Advanced Statistics" course at CU Boulder. But over the course of one semester I quickly found a visceral interest in scientific programming in python (with a few early "easy wins" to boost my confidence). My time in that course culminated in an independent project aimed at a programmatic approach to statistical analysis of scientific data. I chose to sink my teeth into the data from the Kepler Space Telescope, foolishly hoping to discover a new exoplanet in another solar system.

In the process, I learned the basics of object-oriented programming in python, applied computational statistics, data cleaning and processing, exploratory data analysis, model fitting, and data visualization. I used common library packages like matplotlib, numpy, and a few others. While I didn’t discover an exoplanet that the Kepler Science Team didn’t already find, some interesting findings did come from this work. Namely, a binary star system with a highly eccentric companion star (the one orbiting the heavier star) which exhibited a fascinating “elastic bounce” as it suffered the immense tidal forces of the host star – a phenomenon call “star quakes” (like earth-quakes). I followed up with the initial discovery two years later when the signal was scheduled to be detected again, and took observational data from the 24-inch research telescope – which turned out to not be nearly as sensitive as required for such an observation. But at this point, I was already addicted to the process of data-driven discovery.

This repo represents the earliest epoch in my data science saga.