Setting Up a Python Trading Research Environment on Linux

Summary

This tutorial describes an Ubuntu-based setup for Python research using the Anaconda distribution, Conda package management, and isolated virtual environments. It explains that Anaconda bundles common data-analysis libraries and works with Jupyter, while Miniconda is a smaller option that requires installing packages separately. The guide walks through installing Anaconda, initializing its shell behavior, creating an environment with a selected Python version, and adding data-analysis and plotting libraries.

A short example retrieves Apple price data from Stooq with Pandas DataReader, stores it in a DataFrame, and plots adjusted closing prices. The article notes that Yahoo support in Pandas DataReader had changed and uses Stooq instead. It also explains how separate environments help projects use different dependency versions, while mixing Conda and pip packages can complicate later upgrades. The instructions reflect specific historical software versions and Ubuntu steps, so current installers, package availability, and data-source behavior may differ. This is environment and data-access guidance rather than a trading method or performance study.

Key ideas

• Conda environments isolate Python and library versions across research projects.
• Anaconda offers a broad preinstalled package set, while Miniconda reduces the initial footprint.
• The tutorial demonstrates retrieving and plotting Apple market data with common Python analysis libraries.
• Package-manager mixing can make dependency upgrades harder to manage.
• Installation commands and data-source availability are version-dependent and may change over time.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.