Preparing Dow 30 Data for FinRL Stock Trading Models

Summary

This code example demonstrates a data workflow for a FinRL stock-trading project. It fetches a single stock’s price history through both yfinance and FinRL’s Yahoo downloader, then downloads a Dow 30 universe over configured training and trading periods. A feature-engineering step adds technical indicators, VIX data, and a turbulence measure. The workflow then forms date–ticker combinations, joins them with the processed observations, fills missing values with zeros, sorts the panel, splits it into training and trading sets, and saves both sets as CSV files.

The example explains data acquisition and preparation rather than a trading policy or model evaluation. It reports no predictive or investment results. Its output depends on the configured dates, ticker list, indicator settings, and external data sources. Filling missing observations with zero may affect model inputs, and the example does not discuss how to handle market holidays, survivorship bias, data revisions, or leakage between training and trading periods. Those choices should be reviewed before using the prepared data for research.

Key ideas

• The workflow fetches equity prices for a single ticker and a Dow 30 ticker list.
• FinRL feature engineering can add technical indicators, VIX data, and a turbulence measure.
• The processed panel is completed across date and ticker combinations, sorted, and zero-filled.
• The data is split into training and trading periods and saved separately.
• The document demonstrates preparation only and provides no model or strategy performance evidence.

Tags

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