Statis.Co is a Python module encompassing diverse financial tools and functionalities, including indicators, statistical calculations, and connections to data sources. The module is crafted in pure C, seamlessly integrated with Numpy's API, and employs parallelization tools such as OpenMP. The primary goal is to facilitate blazing-fast calculations and empower the module to adeptly handle substantial workloads. To ensure robust memory management and prevent memory leaks, we utilize tools such as Valgrind and Bloomberg's Memray for in-depth analysis of memory usage.
In today's finance industry, there's a growing demand for optimized tools to handle massive workloads swiftly. As financial data complexity rises, cutting-edge technologies play a pivotal role in processing, analyzing, and deriving insights for market analysis, risk assessment, and investment strategies. Staying competitive in finance now hinges on the ability to make informed decisions at unprecedented speeds. Notably, we've achieved over a 70% optimization in execution time compared to pure Python NumPy calculations.
From PyPi. Check the PyPi repo:
pip install statisco
From source:
python setup.py sdist bdist_wheel
pip install .
If you'd like to contribute, please contact me via GitHub.
To compile the C extension, the primary requirements include the installation of Numpy and GCC.
This is a usage example:
from statisco.statistics import closingReturns, mean
msft["MyCloseReturns"] = closingReturns(msft["Adj Close"])
myMean = mean(msft["MyCloseReturns"])
print(f"myMean: {myMean}")
msft.head()
For a more in-depth understanding of usage, refer to the detailed examples provided in the test notebook. We'll be working on developing a documentation site.
This is what I’ll be working on:
- Documentation
- More indicators
- Automatization Classes
- Basic ML models integration
Statis.Co by Hector Miranda is licensed under Attribution-ShareAlike 4.0 International .