With the rise of open source software and open data, there has been increasing interest in open source investment databases and resources. For investment managers and analysts, having access to free, comprehensive databases can provide valuable insights and alpha. In this article, we will explore some of the best open source investment databases available on GitHub and how they can be leveraged.
Compustat and CRSP databases enable fundamental analysis
Two of the most well-known open source financial databases are Compustat and CRSP, which contain fundamental data on publicly traded companies. The Compustat database includes information on companies’ financial statements, while CRSP contains historical stock prices and returns data. By combining these datasets, investors can conduct quantitative analysis and backtesting of fundamental stock-picking strategies. On GitHub, there are projects like ‘crsp_compustat_merge’ that provide convenient access to merged Compustat and CRSP data in CSV format for research purposes.
Alternative data via Quandl and Kaggle
In addition to traditional market data, alternative datasets from sources like Quandl and Kaggle allow investors to tap into new data universes. The Quandl Python package makes it easy to download a wide range of alternative datasets into Pandas DataFrames for analysis. Kaggle offers free access to datasets on topics ranging from housing prices to search trends. By combining alternative data with fundamental data, investors may be able to develop new alpha factors and edges.
Machine learning algorithms and backtesting tools
There are a number of Python libraries focused specifically on finance, including backtesting and machine learning tools. Zipline from Quantopian, PyPortfolioOpt, and Backtrader allow rapid prototyping and backtesting of trading strategies. Scikit-learn, TensorFlow, and Keras provide machine learning capabilities to build predictive signals. With these open source tools, investors can systematically develop and evaluate quantitative investment strategies.
In summary, open source investment databases and tools available on GitHub empower investors with the data and capabilities to discover new signals and anomalies. By combining comprehensive market data with alternative datasets and leading-edge analytics, alpha generation moves into the 21st century.