stock-streams: A quick time series analysis of popular stock tickers using the novel Matrix Profile developed at UCR
Create a virtual environment with the provided requirements file, then simply follow along in the Jupyter Notebook!
The matrix profile efficently annotates a given time series, and using only the MP we can trivially solve a multitude of time series analysis problems including:
- motif discovery
- discords
- regime changes
In this example, we show how we can use the MP on multi-year stock ticker data to visualize common themes and dissonances between two stocks.
These visualizations prove vital for a host of data analysis tasks, and due to efficiency in generating them (as opposed to a brute force pairwise comparison for example), we can utilize the MP in online streaming environments.