GitHub Pages
These are some projects that I have done independently or for consulting work. This page aims to summarize previous projects and link to them to show some of my skillsets in statistics and data science. Everything here was written in R unless indicated otherwise.
This project contains elaborate examples of common statistical models using simulated data from hypothetical studies. Each markdown simulates, visualizes, and models the data. These steps are followed by an easily-understandable interpretation of the results in terms of the study. For more complex designs, the logic behind most steps are also explained. I currently have 5 completed:
This free online book is a continuously updated source of tutorials and background information for analyzing and visualizing fiber photometry time series data from behavioral experiments.
I found some data on COVID-19 grants & loans that were gathered through SEC filings, so I decided to explore and visualize this data to answer some questions I’ve been having. Some of the results so far are surprising - especially how much money 5 airlines were given in grants relative to over 10,000 other companies.
basketball-reference.com has an incredible database of NBA results for both the current and previous seasons. So, I made a Shiny app that is essentially an interactive graph displaying team totals for the 2021 season.
Features:
I stumbled upon a market research scenario prompt that looked like a good way to work on presenting information using different style markdowns. The study design would likely be changed a few times along the way, but this seems like a decent framework for showing knowledge of research design for job applications and the like.
cran version (no longer supported)
github version
The functions in this R packages aim to expedite the data manipulation, exploration, and transformation processes that are useful in working with waveform data. I initially wrote the GCalcium package for a pilot study at University of Colorado when we were testing out GCaMP6 calcium fluorescence imaging, and have since re-written some of the functions for better speed and interpretability. The vignette for the cran version explains the functions in more detail.
These functions were made out of my interest in the stock market. The eventual goal is to re-write and use them to assess the market using 1.) sentimental analysis on news articles about stocks, and 2.) text mining of that company’s financial documents. As of right now, it’s an interactive way of pulling news articles on stocks and the respective company’s financial documents.
These Python functions are essentially shortcuts in connecting to and controlling the BlueStacks android emulator. I wrote these functions so I could compile scripts that automate a few tedious processes in a mobile game I play. The included MSMbot files take advantage of these functions.
I love fantasy baseball and FanGraphs is an incredible source of MLB data, so I wrote a small 3-part tutorial series. The first part teaches writing a relatively simple function that acts as a shortcut to searching on FanGraphs. The second part shows how to gather this data and turn it into a useable format in another function. Last, I give examples of some visualizations of the data and what they mean.
Fantasy football has also been growing on me. This series aims as a way to get R users started in NFL data acquisition, manipulation, and exploration. SQL is also used.
I wanted to compact some of the information that is relevant to my CV but not necessarily common inclusions. This is done through visualizations.