This repository contains my portfolio of investment and finance-related data science projects. These projects are written and presented as Jupyter Notebooks (Python 3), and each link is followed by a short description of project goals, contributions, and dependencies used.
All of the following projects were coded and tested through the Domino Data Lab platform (https://www.dominodatalab.com). I chose to use Domino because the platform allows for easy access to Jupyter notebooks, a built-in terminal, and offline scheduling of jobs for periodic regulation of stock prices.
Better Buy: A Novel Stock Value Joint-Analysis and Comparison Tool
- Performs complete comparisons of competitive strategy, financial strength, growth potential, and valuation of two companies through fundamental and technical analysis.
- Wrote data parser and reader to download 10-Q and 10-K SEC filings and store them in pandas DataFrame, as well as scripts to append certain business performance metrics for trend analysis.
- Designed risk-analysis models using Monte-Carlo Simulations and implemented visualizations using matplotlib and Seaborn.
- Combined data with experts' opinions and shareholder sentiment on business strategy and comapny/management direction to make final "better buy" decision.
Risk/Return Analysis and Predictions for 'FAANG' Stocks
- Here I used pandas Dataframe to represent adjusted closing prices and compute moving averages, daily returns, and associated risk of 'FAANG' tech companies.
- Created visual representations of comparative daily returns and single-stock analysis using Matplotlib, as well as more complex heatmaps and distribution plots using Seaborn.
- Compared and plotted risk of investment between simple return analysis techniques and Monte-Carlo simulations.